Using Qualitative Techniques to Explore Consumer Attitudes: Insights From Group Process Theories

Terry Bristol, Oklahoma State University
Edward F. Fern, Virginia Polytechnic Institute and State University
ABSTRACT - Consumer researchers often use qualitative techniques (e.g., focus groups) to examine consumer attitudes. Yet, our knowledge of the adequacy of these techniques is lacking. This paper examines several theoretical explanations of group phenomena and hypothesizes that group thought eliciting techniques cause attitude polarization. The attitudinal output from these qualitative methods may reflect group attitude formation processes rather than the individual's attitudes.
[ to cite ]:
Terry Bristol and Edward F. Fern (1993) ,"Using Qualitative Techniques to Explore Consumer Attitudes: Insights From Group Process Theories", in NA - Advances in Consumer Research Volume 20, eds. Leigh McAlister and Michael L. Rothschild, Provo, UT : Association for Consumer Research, Pages: 444-448.

Advances in Consumer Research Volume 20, 1993      Pages 444-448


Terry Bristol, Oklahoma State University

Edward F. Fern, Virginia Polytechnic Institute and State University


Consumer researchers often use qualitative techniques (e.g., focus groups) to examine consumer attitudes. Yet, our knowledge of the adequacy of these techniques is lacking. This paper examines several theoretical explanations of group phenomena and hypothesizes that group thought eliciting techniques cause attitude polarization. The attitudinal output from these qualitative methods may reflect group attitude formation processes rather than the individual's attitudes.


The use of qualitative research methods is prevalent in both basic and applied consumer research. For example, the use of focus groups for consumer research has been described as a small and growing industry (McQuarrie and McIntyre 1988). The total expenditure on focus group interviews in the U.S has been estimated to be $312 million (Goldman and McDonald 1987) and is growing. Nevertheless, group interviewing is not enthusiastically embraced by everyone. Clancy and Shulman (1991), the chairman and CEO of Yankelovich Clancy Shulman, claim that focus group interviews are neither serious nor helpful marketing research tools. Moveover, they question whether this qualitative technique is a wise basis for marketing decision making. According to Clancy and Shulman, focus group research is not as accurate or useful as survey research. Nevertheless, their survey of corporate CEO's revealed that 41.2% agreed with the statement "More often than not, the information produced by focus group research is as accurate and useful as the results of survey research at less than half the cost (Clancy and Shulman 1991)." The Clancy and Shulman position may be extreme. Nevertheless, we thought it was important to question whether focus groups, and more generally qualitative research methods provide accurate and useful information about consumers' attitudes.

Two assumptions have motivated this manuscript. First, consumer attitude research is important and is a frequent goal of group qualitative research techniques. Second, since attitude formation is an individual affective and cognitive activity, it is best theorized and studied at the individual respondent level. When the research interest is in group effects on individual attitude formation (e.g., household attitude formation) groups may be appropriate. Because of the inherent conflict between the above two assumptions, this paper will review theories that may explain the extent to which group influences affect individual attitude formation across four different qualitative research techniques.


Qualitative methods are defined in terms of the information gained from their useCsubjective, in-depth understandings of the consumer, and the nature or structure of the consumers' attitudes, feelings, and motivations (Calder 1977; Goldman and McDonald 1987). Qualitative research methods attempt to uncover what people think or how they feel, achieving greater depth and detail of responses, and resulting in close-up descriptions that better realize the subjective nature of the phenomenon studied (Bellenger, Bernhardt, and Goldstrucker 1976; Van Maanen, Dabbs, and Faulkner 1982).

Although, there has been some empirical research on qualitative methods (e.g., Bouchard and Hare 1970; Claxton, Ritchie, and Zaichkowsky 1980; Fern 1982b; Nelson and Frontczak 1988; Reynolds and Johnson 1978), there is a lack of research on what informational output is gained from these methods and how different types of qualitative techniques affect this output.

One type of output frequently sought from the various qualitative techniques is consumer's thoughts, e.g., attitudes, beliefs, opinions, and purchase intentions. For example, these techniques have been used to gain insights into the brand attitudes of women (Kanner 1990), attitudes towards packaging aesthetics (Shapiro 1990), and attitudes towards financial services (Trachtenberg 1987). However, none of the extant research has examined the adequacy of qualitative techniques for exploring attitudes.

There is reason to be concerned about the appropriateness of several of these qualitative techniques for attitude research. In discussing the future of qualitative marketing research, Calder (1977) warned about extending the use of this research approach to all constructs (e.g., attitudes, values, traits, roles, norms, etc.), without regard to their amenability to existing scientific methods. Axelrod (1975) specifically warned against the use of focus groups to determine consumers' preferences: "This kind of 'voting' for preferences is more often than not just an intellectual exercise (p.6)." Dietz (1975) represents the most extreme position on the appropriateness of focus groups for attitude research. Dietz claimed that focus groups are sold at inflated prices "because they are sold under the pretense of providing insight into consumer concepts, perceptions, opinions, and attitudes. It is a pretense that in many instances has deceived the research supplier more than the receiver (p. 6)."

To illustrate the nature of the problem, consider the following. One authority on focus groups cited a case where a respondent declared that he would not buy a prestige car because it was ostentatious. After one hour and thirty minutes of focus group discussion, the same respondent declared the he was fickle and that he would buy a prestige car and show it off to his friends. Presumably, this revelation provides insights into the respondent's "true" attitude about prestige cars. However, it may also indicate that people change their attitudes during group discussion. At best, professional judgment by the moderator is necessary to make these types of determinations.

In summary, our theoretical and empirical knowledge about qualitative techniques is lacking. As a first step in filling this void, we reviewed the group process literature to uncover theoretical explanations for differences in the effectiveness of qualitative techniques for measuring consumer attitudes. The implications of this review for the conduct of attitude research will be discussed. First, however, we will review the types of qualitative techniques used for examining and assessing consumer attitudes.

Qualitative Research Techniques

Consumer researchers have reported the use of four types of qualitative research techniques to investigate consumer attitudes: open-ended surveys, individual interviews, focus group interviews, and nominal group interviews. These techniques vary in the exact procedures and settings used to collect the information, and in their assumptions of the quality of the output gained.

Open-ended surveys (or self-administered questionnaires) provide respondents with written instructions and questions, yet do not provide structured responses. Researchers often use surveys to measure consumer attitudes. Surveys are easily and economically administered on a mass basis to a sample of individuals. Additionally, researchers often use unstructured, open-ended response formats to preclude constraining and/or biasing individual's responses (Schuman and Presser 1981).

Individual interviews involve relatively open-ended, but interviewer guided, discussions of specific topics. Interviews allow the researcher as interviewer to probe responses and redirect questions towards the respondent. Thus, individual interviews can be more adaptive and less structured than surveys. However, the former are costly when a large number of respondents are desired.

Like individual interviews, focus group interviews also involve a moderator guided open-ended discussion of specific topics. However, focus groups are interactive as the topics are simultaneously discussed by a small group of individuals. Researchers often use focus groups to gain insight into consumer attitudes because of the convenience in interviewing several respondents simultaneously, and because consumers are more likely to respond in a group interview setting (Fern 1982a). Thus, focus groups can provide data quicker and more economically than individual interviews (Stewart and Shamdasani 1990).

Nominal group techniques are highly structured group interviews that restrict spontaneous interaction (Claxton, Ritchie, and Zaichkowsky 1980; Stewart and Shamdasani 1990). Nominal groups differ from focus groups in the added structure to the interview. For example, participant's responses are often written and shared with the group on a cumulative basis rather than spontaneously and openly discussed as found in focus groups. Additionally, vocal evaluation of the output is prohibitedCthe moderator collects written evaluations by each individual. Like focus groups, the nominal group technique purportedly results in higher involvement and response rates compared to non-group techniques (Claxton, Ritchie, and Zaichkowsky 1980).

Thus, the users of each technique have assumed that they are valid and useful methods for examining consumer attitudes. Current theories in social psychology can clarify the presumed attitudinal output of each of these techniques.


Although the effects of using qualitative techniques on the quality of attitudinal output has not been empirically examined, some work has proceeded in social psychology that applies to this problem. Many studies have found that initial tendencies of individual group members intensify or sometimes change with group discussion (Kaplan 1987; Whitney and Smith 1983). Several theories have been suggested to explain this phenomenon, several of which apply to the use of qualitative techniques to measure consumer attitudes. Reviewed below are theories of social facilitation, social impact, social comparison, and persuasive-arguments.

Social Facilitation and Social Impact

Social facilitation is perhaps the simplest theory from which predictions can be posited about measuring attitudes across social and nonsocial settings. This theory suggests that the mere presence of others is a sufficient explanation of behavior in groups (Allport 1924). The presence of observers or coactors has been found to result in greater individual effort and performance (Markus 1978; Zajonc 1980). However, we are interested in the effect of others on individual attitudes. Ideas from self-awareness theory (Duval and Wicklund 1972) apply here. The presence of others leads individuals to focus attention on themselves and increases self-awareness and thought about one's own attitudes and feelings (Carver 1979). Increased self-awareness has been found to lead to polarized attitudes and evaluations (Ickes, Wicklund, and Ferris 1973; Scheier and Carver 1977). Attitude polarization refers to individuals adopting more extreme attitudinal positions than those previously heldCit is a shift or change in degree but not in direction (Allison and Messick 1987). Thus, the mere presence of others leads to increased self-awareness that increases thought about one's own attitudes, resulting in attitude polarization.

Focus group and nominal group interviews contain 5 to 12 participants, whose presence may lead to increased self-awareness and attitude polarization by each individual. The interviewer is present in an individual interview and thus, this technique also may lead to polarization. However, no other individual need be present for an individual to respond to an open-ended survey.

In summary, we expect individuals' attitudes to become more polarized when others are present during the thought eliciting procedure. Therefore, we expect that the interviewer-directed thought eliciting methods (i.e., focus groups, nominal groups, and individual interviews) will produce more polarized attitudes than using open-ended, self-administered questionnaires.

Any polarization in the individual's attitudes that is detected in the thought eliciting procedure should endure only while the individual is subject to the self-awareness and increased thinking about her or his position that results from the presence of others. When others are not present, individuals will fall back on their pre-existing attitude state. The polarized beliefs will not be salient or persist beyond the interview setting. The polarization of attitudes that is due to self-awareness has been found to dissipate over time (Ickes, Wicklund, and Ferris 1973). Therefore, we have concluded that any attitude polarization that results from the presence of others will be temporary and will not endure beyond the interview situation.

Social impact theory goes beyond the mere presence of others in its predictions. This theory predicts that the impact of others on individuals' beliefs, cognitions, values, and emotions is some power function of the number of other people present (LatanT and Nida 1980). LatanT and Nida (1980) provide evidence from the conformity literature in support of their notions. According to social impact theory the effects of social facilitation should increase with the presence of more individuals. Thus, the greater the number of other people present, the greater self-awareness and the more polarized the attitudes. LatanT and Nida would also predict that the status of group members would impact on attitude polarizationChigher status individuals would cause greater attitude shift than lower status individuals.

Alternate predictions of attitude polarization due to mere presence result from applying social impact theory. Attitude shifts should occur in the group techniques and these shifts should be greater than in individual interviews in which an interviewer is present with the respondent (i.e., a group of two). Moreover, any attitude shift in individual interviews should be greater than shifts in open-ended surveys and could be attributed to the mere presence of the interviewer. Again, attitude polarization due to the impact of the social situation should endure only as long as the individual remains in the presence of others.

In summary, individual attitude polarization should increase as the number of others present increases such that polarization in focus and nominal groups will be greater than that in individual interviews, which will be greater than that in surveys. Additionally, as the number of individuals within the each group increases the positive effect on attitude polarization should increase. Finally, the shift effect should be transitory. Attitude polarization resulting from the presence of others is temporary and it should not endure beyond the interview situation.

Normative Influence Through Social Comparison

Social comparison theory was originally formulated by Festinger (1954) to explain the effects of social communication on opinion change in groups. One derivation of the theory has been used to go beyond mere presence theories to explain the polarization of group member attitudes that result from group discussion. Simply, group members' desires to be favorably evaluated leads them to adopt an attitude that is more extreme than the group norm, to the extent that they are aware of this normative position (Goethals and Zanna 1979).

There are three necessary conditions for social comparisons to cause attitude polarization: (1) participants must desire to be favorably evaluated, (2) the setting must provide a standard of comparison, and (3) the setting must allow for the evaluations of others. Thus, in settings where others provide a standard of comparison, a desire to be favorably evaluated may motivate individuals to adopt/express more extreme attitudes (Allison and Messick 1987; Harkins and Szymanski 1987). The standard of comparison in group interviews is a normative position on an issue that is explicitly stated by the interviewer/researcher or implicitly derived from the statements provided by group members during the discussion. Additionally, to be evaluated (favorably or unfavorably), the individual's views must be expressed and identified to others (Harkins and Szymanski 1987).

The desire to be favorably evaluated by others is pervasive. This desire is a source of evaluation apprehensionCan individual's anticipation of positive or negative outcomes when around others serves as an incentive function (Cottrell 1972). Similarly, the impression management literature has assumed that a primary goal of self-presentation is the attainment of social approval (Arkin 1981). According to social impact theory, we might expect this desire for favorable evaluation to be greater when other group members are acquaintances rather than strangers (i.e., people they will never see again) and higher status individuals rather than lower status individuals.

Standards of comparison can be personal, objective, or social (Harkins and Szymanski 1987). It is this latter social standard that is produced in a setting in which others' views are shared. Individuals have been found to compare themselves with similar others in these settings (Zanna, Goethals, and Hill 1975). Both objective and social standards have been found to affect various types of responses (Harkins and Szymanski 1987).

Applied to the context at hand, the three necessary conditions for social comparisons to cause attitude polarization are only present when focus group interviews are used. Standards for comparison can develop in both focus groups and nominal groups, but are less likely to develop in individual interviews. In focus groups each individual's attitudes are expressed and can become the focal point of discussion during the group exchange. However, in nominal groups the attitudes of individuals may be elicited but no conscious effort is made to discuss or otherwise scrutinize individuals' contributions. Since the ideas expressed within nominal groups are not openly discussed, the individual participant is less likely to be apprehensive about being evaluated by the other group members. With less emphasis on the evaluation of others, standards of comparison are less likely to affect attitude formation and change in nominal groups. The result should be less shifting in attitudes compared to focus group interviews.

In individual interviews and open-ended surveys, other members are not available for evaluation, consequently no standard of comparison exists, assuming the interviewer or questionnaire does not offer cues to such a standard. Thus, attitudes expressed by participants in individual interviews and open-ended surveys should not polarize or shift from what the individual believed prior to the interview or questionnaire, compared to focus group interviews.

In summary, individual attitude polarization should increase under normative pressures. Polarization should be greater among focus group than nominal group participants. As previously noted, group attitude polarization should be more pronounced than polarization in individual interviews and open-ended surveys.

Any change in the individual's attitudes that is detected in the interview will endure only as long as normative pressures are acting on the individual. When the norms are no longer governing individuals' behaviors, they will fall back on their pre-existing attitude structures.

Informational Influence Through Persuasive Arguments

An alternate explanation of attitude polarization or change in groups is information influence or persuasive-arguments theory (Kaplan 1987; Vinokur and Burnstein 1974). This theory posits that the exchange of information in groups can lead members to consider facts that they had not previously considered when initially forming their attitudes (Allison and Messick 1987). This new information could not only lead to attitude polarization, but to attitude change or depolarization (Vinokur and Burnstein 1978). Weakly held attitudes may be more easily changed because the individual is more influenced given contradictory information provided by others. Evidence for both polarization and depolarization given group discussion has been reported (Vinokur and Burnstein 1978; Whitney and Smith 1983).

Information is shared in both focus groups and nominal groups but is not provided to respondents in individual interviews and open-ended surveys. In focus groups each individual contributes information about her or his feelings. This information is often the focal point of group discussion. Although not interactively discussed in the group, information about the attitude object is also expressed in nominal group interviews. The individual participant is the only source of information in individual interviews and open-ended surveys, assuming the interviewer or questionnaire produce no demand artifacts. Therefore, attitude polarization should be greater in group interviews because more beliefs, opinions, and feelings are shared and discussed among participants.

As opposed to the other theories presented, information influence should result in an enduring change in the individual's attitudes. Attitudes that change because of new or additional information should remain stable outside the interview setting. The new information or beliefs should be salient and persist beyond the interview.

To summarize, individual attitude change or polarization, resulting from shared information, will be greater for group interview techniques than individual interviews or open-ended surveys. Moreover, the attitude change or polarization resulting from information influence is relatively permanent and will endure beyond the interview situation.


The Table summarizes the relationships between the techniques and degree of attitude shift predicted by the theories. If attitudes are influenced by the group's or interviewer's presence, then the interview technique itself may unintentionally produce polarization. Similarly, if attitudes are influenced by group norms, or information shared in the group, then group qualitative methods for collecting attitudinal information may inadvertently polarize or change individual members' attitudes. Regardless, it is not known whether these attitude shifts are transitory or enduring. Individuals may express attitudes during a discussion that differ from their previous attitudes. After the discussion is completed, individuals may either revert to their original attitudes or permanently adopt the polarized attitudes. In either case, the output may result from group attitude formation processes rather than individual attitude processes. Consequently, managerial prescriptions and theoretical descriptions based on group attitude research procedures may be problematic. If the attitudes are transitory, prescriptions will be based on attitudes that no longer exist. If the new attitudes are enduring, the prescriptions will only apply to those experiencing comparable group phenomena.




Several theories were reviewed which may add to our understanding of the use of qualitative techniques. Other theories of group processes can be applied to this context. For example, coalition formation (Goldman and McDonald 1987) and social identity theory (Tajfel and Turner 1986; Mackie 1986) have been used to explain the polarization of attitudes within groups. However, these offer no further insight into the problem because predictions derived from these theories duplicate those posited above.

We hope that this paper will prompt empirical exploration of the efficacy of using qualitative techniques to examine consumer attitudes. While much has been written about qualitative research, most of what is known about these techniques is based on experiential reports by moderators and interviewers. This information may be useful but it tends to be idiosyncratic to the researcher and her or his firm. Therefore, empirical validation is necessary to discover whether the output from group qualitative techniques represents group attitude formation processes or individual attitudes. If no attitude shift is found, we can assume that the output represents individual attitudes and not the effects of group interaction. Additionally, we need to show empirically the utility of these qualitative techniques for exploring consumer attitudes.

Empirical testing of these notions will enable prescriptions about collecting attitude information without the unintended effects of group interaction. If we find that attitudes do not change or polarize in the group techniques, then prescriptions for choosing among techniques can be based on factors such as cost and convenience, rather than on quality of the output. Considering the frequency of use and the amount of money spent on qualitative research, the results of further empirical studies should be highly relevant to both academicians and practitioners.


Allison, Scott T. and David M. Messick (1987), "From Individual Inputs to Group Outputs, and Back Again: Group Processes and Inferences About Members," in Group Processes, ed. Clyde Hendrick, Newbury Park, CA: Sage Publications, 111-143.

Allport, Floyd Henry (1924), Social Psychology, Boston, MA: Houghton Mifflin.

Arkin, Robert M. (1981), "Self-presentation Styles," in Impression Management: Theory and Social Psychological Research, ed. James T. Tedeschi, New York: Academic Press, 311-333.

Axelrod, M. D. (1975), "Marketers Get an Eyeful When Focus Groups Expose Products, Ideas, Images, Ad Copy, Etc. to Consumers," Marketing News, 9 (February 28), 6-7.

Bellenger, Danny N., Kenneth L. Bernhardt, and Jac L. Goldstrucker (1976), Qualitative Research in Marketing, Chicago, IL: American Marketing Association.

Bouchard, Thomas J., Jr., and Melanie Hare (1970), "Size, Performance, and Potential in Brainstorming Groups," Journal of Applied Psychology, 54 (February), 51-55.

Calder, Bobby J. (1977), "Focus Groups and the Nature of Qualitative Marketing Research," Journal of Marketing Research, 14 (August), 353-364.

Carver, Charles S. (1979), "A Cybernetic Model of Self-Attention Processes," Journal of Personality and Social Psychology, 37 (August), 1251-1281.

Clancy, Kevin J. and Robert S. Shulman (1991), The Marketing Revolution, New York: Harper Business.

Claxton, John D., J. R. Brent Ritchie, and Judy Zaichkowsky (1976), "The Nominal Group Technique: Its Potential for Consumer Research," Journal of Consumer Research, 7 (December), 308-313.

Cottrell, Nickolas B. (1972), "Social Facilitation," in Experimental Social Psychology, ed. C. G. McClintock, New York: Holt, Rinehart, and Winston, 185-236.

Dietz, Leonhard (1975), "Can Focus Group Interviews Survive?" Marketing News, 9 (October 10), 6-7.

Duval, Shelley and Robert A. Wicklund (1972), A Theory of Objective Self-Awareness, New York: Academic Press.

Fern, Edward F. (1982a), "Why Do Focus Groups Work: A Review and Integration of Small Group Process Theories," in Advances in Consumer Research, Vol. 9, ed. Andrew Mitchell, Ann Arbor, MI: Association for Consumer Research, 444-451.

Fern, Edward F. (1982b), "The Use of Focus Groups for Idea Generation: The Effects of Group Size, Acquaintanceship, and Moderator on Response Quantity and Quality," Journal of Marketing Research, 19 (February), 1-13.

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

Goethals, George R. and Mark P. Zanna (1979), "The Role of Social Comparison in Choice Shifts," Journal of Personality and Social Psychology, 37 (September), 1469-1476.

Goldman, Alfred E. and Susan Schwartz McDonald (1987), The Group Depth Interview: Principles and Practice, Englewood Cliffs, NJ: Prentice Hall, Inc.

Harkins, Stephan G. and Kate Szymanski (1987), "Social Loafing and Social Facilitation: New Wine in Old Bottles," in Group Processes and Intergroup Relations, ed. Clyde Hendrick, Newbury Park, CA: Sage Publications, 167-188.

Ickes, William John, Robert A. Wicklund, and C. Brian Ferris (1973), "Objective Self-Awareness and Self-Esteem," Journal of Experimental Social Psychology, 9 (May), 202-219.

Kanner, Bernice (1990), "The Secret Life of the Female Consumer," Working Woman, 15 (December), 68-71.

Kaplan, Martin F. (1987), "The Influencing Process in Group Decision Making," in Group Processes, ed. Clyde Hendrick, Newbury Park, CA: Sage Publications, 189-212.

LatanT, Bibb and Steve Nida (1980), "Social Impact Theory and Group Influence: A Social Engineering Perspective," in Psychology of Group Influence, ed. Paul B. Paulus, Hillsdale, NJ: Lawrence Erlbaum, 3-34.

Mackie, Diane M. (1986), "Social Identification Effects in Group Polarization," Journal of Personality and Social Psychology, 50 (April), 720-728.

Markus, Hazel (1978), "The Effect of Mere Presence on Social Facilitation: An Unobtrusive Test," Journal of Experimental Social Psychology, 14 (July), 389-397.

McQuarrie, Edward F. and Shelby H. McIntyre (1988), "Conceptual Underpinnings for the Use of Group Interviews in Consumer Research," in Advances in Consumer Research, Vol. 15, ed. Michael J. Houston, Provo, UT: Association for Consumer Research, 580-586.

Nelson, James E. and Nancy T. Frontczak (1988), "How Acquaintanceship and Analyst Can Influence Focus Group Results," Journal of Advertising, 17 (1), 41-48.

Reynolds, Fred D. and Deborah K. Johnson (1978), "Validity of Focus Group Findings," Journal of Advertising Research, 18 (June), 21-24.

Scheier, Michael F. and Charles S. Carver (1977), "Self-Focused Attention and the Experience of Emotion: Attraction, Repulsion, Elation, and Depression," Journal of Personality and Social Psychology, 35 (September), 625-636.

Schuman, Howard and Stanley Presser (1981), Questions and Answers in Attitude Surveys, San Diego, CA: Academic Press.

Shapiro, Sid (1990), "Focus Groups: The First Step in Package Design," Marketing News, 24 (September 3), 15-17.

Stewart, David W. and Prem N. Shamdasani (1990), Focus Groups: Theory and Practice, Newbury Park, CA: Sage Publications.

Tajfel, Henri and J. C. Turner (1986), "An Integrative Theory of Intergroup Relations," in The Psychology of Intergroup Relations, eds. Stephen Worchel and William G. Austin, Chicago, IL: Nelson Hall, 7-24.

Trachtenberg, Jeffrey A. (1987), "Listening, the Old-Fashioned Way," Forbes, 140 (October 5), 202-204.

Van Maanen, John, James M. Dabbs, Jr., and Robert R. Faulkner (1982), Varieties of Qualitative Research, Beverly Hills, CA: Sage Publications.

Vinokur, Amiram and Eugene Burnstein (1974), "Effects of Partially Shared Persuasive Arguments on Group-Induced Shifts: A Group-Problem-Solving Approach," Journal of Personality and Social Psychology, 29 (March), 305-315.

Vinokur, Amiram and Eugene Burnstein (1978), "Depolarization of Attitudes in Groups," Journal of Personality and Social Psychology, 36 (August), 872-885.

Whitney, John C. and Ruth A. Smith (1983), "Effects of Group Cohesiveness on Attitude Polarization and the Acquisition of Knowledge in a Strategic Planning Context," Journal of Marketing Research, 20 (May), 167-176.

Wicklund, Robert A. (1975), "Objective Self-Awareness," in Advances in Experimental Social Psychology, Vol. 8, ed. Leonard Berkowitz, New York: Academic Press, 233-276.

Zajonc, Robert B. (1980), "Compresence," in Psychology of Group Influence, ed. Paul B. Paulus, Hillsdale, NJ: Lawrence Erlbaum, 35-60.

Zanna, Mark P., George R. Goethals, and Janice F. Hill (1975), "Evaluating a Sex-related Ability: Social Comparison with Similar Others and Standard Setters," Journal of Experimental Social Psychology, 11 (January), 86-93.