Special Session Summary Word-Of-Mouth Behaviour: What Motivates It? How Does It Affect Product Evaluation?

Citation:

Cindy M.Y. Chung (2003) ,"Special Session Summary Word-Of-Mouth Behaviour: What Motivates It? How Does It Affect Product Evaluation?", in E - European Advances in Consumer Research Volume 6, eds. Darach Turley and Stephen Brown, Provo, UT : Association for Consumer Research, Pages: 235-239.

European Advances in Consumer Research Volume 6, 2003      Pages 235-239

SPECIAL SESSION SUMMARY

WORD-OF-MOUTH BEHAVIOUR: WHAT MOTIVATES IT? HOW DOES IT AFFECT PRODUCT EVALUATION?

Cindy M.Y. Chung, Nanyang Technological University, Singapore

SESSION OVERVIEW

Word-of-Mouth (WOM) is the informal communication between private parties concerning the evaluation of goods and services (Westbrook 1987; Singh 1988). It is a key marketing phenomenon that has been found to facilitate the sales of a wide range of products including professional services (Smith and Meyer 1980), movies (Mizerski 1982), automobiles (Swan and Oliver 1989), travel destinations (Gitelson and Crompton 1983), and innovations (Rogers 1983). Given the powerful impact of WOM, companies can greatly benefit from knowing how to encourage consumers to give WOM, and what effects WOM messages have on WOM recipients. This session focuses on adding to the current understanding of WOM motivation, and the impacts of WOM. It includes three papers. The first two examine factors that motivate WOM behaviour. The third paper analyses how the presentation order of positive and negative information on a product within a WOM message can affect the recipient’s product attitude and purchase intention.

It has been commonly assumed that positive words come from consumers who like a product. However, product attitudes do not necessarily lead to WOM behaviour (Swan and Oliver 1989). Consumers may need to be motivated to give WOM, in addition to having positive opinions about the product. Relatively little research has been conducted on the motivation of WOM. Some existing studies identify factors that motivate WOM but do not systematically test them (e.g., Dichter 166). Also, the effectiveness of some popular WOM-incentive programmes for inducing WOM behaviour, such as recommend-a-friend programmes, has not been examined. The first two papers address these concerns.

In the first paper, Chung argues that by increasing personal relevance of giving WOM, marketers can encourage more WOM on products that consumers already like. The association between products and the self-concept, or self-relatedness, encourages consumers to give WOM for self-promotion purpose (Arndt 1967). In a series of studies, Chung finds a significant effect of self-relatedness on the amount of detail given in WOM, and the valence of WOM. Chung further explains how product ownership moderates the effects of self-relatedness on WOM valence. When the WOM giver owns the product, self-relatedness leads him or her to give more positively valenced WOM. This indicates an attempt to put himself or herself in a better light by exaggerating consumption experiences. Finally, Chung shows that when expecting a social evaluation on the WOM, the WOM giver talks more about the product he or she likes, and is more prone to exaggerating positive consumption experiences. In sum, this research shows that marketers can increase personal relevance of giving WOM by leading consumers to associate products with some important aspects of the self-concept, hence stimulating more WOM on these products.

In the second presentation, Chew, Tambyah, and Wirtz investigate consumer opinions on WOM-incentive programmes such as recommend-a-friend programmes, and their effectiveness. Through semi-structured interviews with consumers, the authors identify those who express either positive or negative attitudes towards incentive programmes, and obtain explanations for their attitudes. For consumers who view these programmes positively, incentives strongly motivate recommendation behaviour, particularly among strong-tie consumers. The motivation to recommend is even stronger when both the WOM giver and recipient receive incentives, thus creating a win-win situation. The authors also classify different commonly used incentive schemes based on data from the interviews, and discuss their popularity and appropriateness. Overall, the research by Chew et al. provides useful managerial insights into what types of incentive scheme to adopt for improving the effectiveness of WOM-incentive programmes.

Apart from being able to encourage WOM, marketers might find it useful to know the effects of WOM messages on the subsequent product evaluation and decision. WOM often has a greater impact on consumers than media-based communications, because it is considered more reliable and trustworthy (Day 1971). One survey, for instance, found that 60% of respondents bought products solely on the basis of personal recommendation, whereas only 29% suggested they had ever bought a product solely due to a TV advertisement (Globe and Mail 1999). Generally, positive WOM leads to positive product evaluations, and negative WOM, negative evaluations. Whereas the impact of either mainly positive or mainly negative WOM on product evaluation is well documented (Herr, Kardes, and Kim 1991; Wilson and Peterson 1989), the impact of mixed WOM is not. The effect of messages containing both positive and negative information has by and large been studied only in the psychology domain. It is pertinent to study this effect in the context of WOM because WOM messages frequently contain both positive and negative information. To the extent that the two types of information may affect product evaluations in opposite directions, different presentation orders of the information within the same WOM message may affect the eventual consumer decision. The final paper in this session examines this issue.

Ho and Chung report the results from two experiments that test the effects of mixed WOM on product evaluation. They vary the presentation order of positive and negative WOM from a single source on a restaurant, and measure product attitude and purchase intention afterwards. They find that regardless of subject involvement, the information appearing later in the message has a greater impact on product evaluation than that appearing earlier. The resuts differ from previous findings that demonstrated a recency effect only under low involvement conditions. The authors provide insights into how the WOM giver can arrange the order of positive and negative information in mixed WOM to achieve a maximum impact on the recipient.

Taken together, the three presentations in this session advance the understanding on WOM behaviour, which has been so far fairly under-researched. One conclusion is that marketers can motivate WOM to their advantage. However, the quality of products and services that they subsequently provide must live up to the WOM communicated. Otherwise, any overly positive WOM will backfire. Another insight is that since consumers may have personally relevant goals or incentives in mind while giving WOM, the resulting WOM may not always be as accurate and credible as it has been commonly believed. Finally, if a simple reordering of information within a WOM message can change product evaluation, avid WOM givers such as opinion leaders and market mavens can improve their persuasive power by strategically arranging the content of their WOM messages.

ABSTRACT - S

 

AN INVESTIGATION ON HOW PERSONAL RELEVANCE LEADS TO GREATER AMOUNT OF WORD-OF-MOUTH AND MORE POSITIVELY VALENCED WORD-OF-MOUTH

Cindy M.Y. Chung, Nanyang Technological University, Singapore

Most marketers agree that positive word-of-mouth (WOM) helps sales and builds reputation. However, many also wonder how they can encourage positive words about their products or services. This three-study research examines factors that may motivate satisfied consumers to give WOM. First, products that are associated with some aspect of the self-concept are expected to motivate WOM better than those that do not. Study 1 examines this association. Second, ownership of products may moderate the impacts of this association on WOM. Study 2 tests for this moderation. Finally, an expectation of social evaluation on WOM may also motivate consumers to give WOM. Study 3 reports evidence on this prediction.

The author argues that when consumers find personal relevance in giving WOM, they will be more willing to provide detail on products they like. In the realm of social judgement and decision-making, personal relevance is regarded as one of the most important motivational variables (Petty and Cacioppo 1986). An issue or behaviour is personally relevant when it is related to important goals, or when it bears significant consequences. Personal relevance motivates a person to spend time and effort on information processing. He or she will attend to information that is relevant to the current issue, and try to comprehend and elaborate on it. The goal is to arrive at accurate and objective conclusions or behaviour. The information-processing goal can sometimes be directional in nature, when the information processor engages in defense or impression management (Chaiken, Giner-Sorolla, and Chen 1996; Kunda 1987). He or she will select or distort information for the purpose of defending the self against attacks, or for portraying a positive impression to others. Thus, directional goals often result in biased judgement and behaviour.

Translated into the context of WOM, personal relevance may prompt the WOM giver to spend relatively more time and effort on communicating product information. As such, WOM is expected to contain more detail. Additionally, when the goal is to promote the self-concept (Arndt 1967), or to show superiority (Dichter 1966), the WOM giver may be biased towards disseminating selective or even distorted product detail for the purpose of achieving his or her directional goal. Therefore, if the WOM giver has a positive attitude towards a product, he or she may be motivated to exaggerate the favourable aspects of the product. Given everyday discourse is frequently centered on products (Richins 1994), an association between a product and an important aspect of the self-concept should help to motivate self-promotion driven WOM behaviour. Furthermore, consumers regard some products as their "extended-self," and a means to express personal values (Belk 1988). Ownership of a product is therefore expected to enhance a self-promotion goal, and moderate the effects of the product-self association on WOM behaviour. Finally, social feedback is an important source of self-esteem (Baumeister and Jones 1978). To the extent that the WOM giver likes a product and the product associates with his or her self-concept, an anticipation of social evaluation on WOM will encourage self-promotion, and lead to more WOM detail, and more exaggeration of positive consumption experiences.

Study 1 examined the effects of a product-self association on total WOM (i.e., the number of product-related details) and WOM valence (i.e., the difference between the number of positive and the number of negative evaluations on the product). Subjects were randomly assigned to either of the conditions in a one-factor (product type: self-related vs. not self-related) between-subjects design. Probed by a trained confederate, subjects gave WOM on either type of products towards which they have earlier expressed positive attitudes. More total WOM was given on products that were self-related than not (’s=13.42 vs. 7.78; t(66)=2.44, p<.05). However, WOM valence did not differ across the two conditions (p=.15). Subjects gave relatively more of both positive and negative evaluations on self-related products, resulting in no difference in the valence between the two conditions.

Study 2 measured the same dependent variables, using a survey methodology that included a 2 (product type: self-related vs. not) by 2 (ownership vs. no ownership) between-subjects design. Total WOM on self-related (vs. not self-related) products was again greater ( ’s=7.35 vs. 5.48; F(1, 65)=7.23, p<.01). When subjects owned the products, those that were self-related (vs. not) inspired greater WOM valence ( ’s=3.94 vs. 1.88; F(1,65)=3.92, p=.05). Unexpectedly, when subjects did not own the products, those that were not self-related (vs. those that were) inspired marginally greater WOM valence ( ’s=2.78 vs. 0.77; F(1,65)=3.85, p=.06).

The final study focused on the effects of social evaluation on the same dependent variables. Subjects were randomly assigned to one of the conditions in a 2 (social evaluation: expected vs. not expected) by 2 (advertising copy: self-related vs. not self-related) between-subjects design, and gave WOM on a bogus brand of snack mixes. Half of them were told that a panel of food experts would judge their WOM. The other half did not expect this evaluation. Half of them also read ad copy that associated the snack mixes to showing good taste (i.e., a desirable aspect of the self-concept as found in a pretest); half read ad copy without this association. The results showed that subjects who expected evaluation gave marginally more total WOM than those who did not( ’s=10.82 vs. 8.97; F(1,65)=2.95, p=.09). An analysis contrasting the not-expected/ not-self-related group ( =.89) with the other three conditions partially confirmed that WOM valence was greater when the ad copy associated the snack mixes to the self-concept ( =2.76), when subjects expected evaluation on their WOM opinions ( =3.44), or both ( =1.94) (F_Y(1,65)=3.37, p=.07).

Conceptually, this research offers initial evidence on how motivational goals such as self-concept promotion may encourage the WOM giver to give more WOM on products that they already like, and to even exaggerate this liking, resulting in more positively valenced WOM. Marketers may stimulate WOM to their advantage by appropriately associating some products to important aspects of the self-concept.

 

THE ROLE OF INCENTIVES ON RECOMMENDATION BEHAVIOUR

Patricia Chew, National University of Singapore, Singapore

Siok Kuan Tambyah, National University of Singapore, Singapore

Jochen Wirtz, National University of Singapore, Singapore

Firms have been increasingly using word-of-mouth (WOM) incentive programmes, such as recommend-a-friend programmes, in the hope of increasing their customer base. However, the effectiveness of such programmes has not been extensively investigated. Using semi-structured interviews, consumers’ opinions about these programmes, as well as their non-incentivised and incentivised WOM behaviour after a positive product or service experience, were examined.

Preliminary results show that respondents fall into two groups with regard to their beliefs and attitudes towards existing WOM incentive programmes. About one-third of the respondents had a positive view of the programmes. The reasons are two-fold. First is the utility to the WOM giver. The WOM giver receives rewards for making recommendations and these serve as motivators, encouraging more of the same behaviour. Second is the utility to the WOM recipient. Some respondents felt that if WOM recipients purchased the products and/or services, they would fulfil a need and enjoy the benefits accruing from their purchase.

In contrast, about two-thirds of the respondents viewed current WOM incentive programmes negatively and refrained from participating in them. The reasons can be classified into programme characteristics and personal characteristics. For programme characteristics, there were three aspects that respondents disliked. The first was the respondents’ perception of why companies initiated these programmes. Respondents viewed these programmes as another "sales ploy" to attract customers at times when business was poor. The second aspect was the programme design. Some programmes required respondents to talk to their friends and obtain a certain number of names. This was viewed as a hassle and an inconvenience because of the time and effort needed to comply with such WOM requirements. Although some other programmes only required respondents to simply fill in a list of their friends’ contact details, it was felt that their friends would not welcome unsolicited contact by the firm. Third, the types of rewards were not attractive. For example, the reward was a product the respondents did not need or want, or the absolute amount of the rewards was too paltry.

For personal characteristics, there were three reasons why respondents were disinterested in the programmes. The first was what motivated their WOM behaviour. One motivator was the good product or service experience that respondents wanted to share with others. Another motivator was altruism. Respondents wanted to help the WOM recipient, with no motive for personal gain. Second, some respondents (e.g., working adults) were too busy to act on the incentives. Third, some respondents were simply not deal-prone, or they felt that they did not have the "pushy salesman" kind of character. So, they did not participate in the programmes.

For the respondents who held a positive view of current programmes, incentives were strong motivators for them to make recommendations, but the incentive-WOM relationship was moderated by tie strength. The non-incentivised WOM behaviour of these respondents can be divided into three categories: those who normally did not talk about their positive experiences; those who would only share spontaneously with strong ties, but needed to be prompted before they would share with weak ties; and finally, those who shared with both strong ties and weak ties, when prompted by conversations or direct questions. In general, if respondents did not at first give WOM, incentives motivated them to share with strong ties, but not weak ties. If they previously only shared with strong ties, incentives motivated sharing with weak ties as well. If they previously needed to be prompted to share, incentives motivated spontaneous sharing.

The discussion has so far been focused on programmes that awarded incentives to only the WOM giver. If incentives were also given to the WOM recipient, the majority of the respondents who showed a favourable opinion of current programmes would make more effort to recommend. These respondents felt that incentives given to both parties represented a win-win siuation. It would be easier for them to make recommendations because they would not be the only ones benefiting. Also, the WOM recipient would be motivated to sign up or buy, since they would immediately receive a reward. However, more effort put in did not necessarily mean the WOM giver would tell more people. Some might, but others might try harder to persuade the same group of people they would have told when the incentive was only given to the WOM giver.

For respondents who had a negative opinion about current programmes, an incentive that was of the desired kind would motivate only a small number to become positively enthusiastic about making recommendations. Nearly two-thirds of the group would talk to others, but would not put in too much effort, or go out of their way to recommend in order to receive an incentive. Most of them reiterated that they were motivated more by altruism, than by incentives that were "immaterial." About nearly one quarter of this group would still not be motivated by the various forms of incentives. However, if incentives were given to both the giver and the recipient, some of those who were less enthusiastic about incentivised WOM would be more positive about making recommendations.

As can be seen, having the desired kind of incentives can motivate incentivised WOM. The desired kind of incentives as suggested by the respondents can be classified into four types. The first was cash, and the second was products or services related to their purchase, like discount vouchers, cash rebates, and free meals. The third kind was desirable hedonic products, which may be unrelated to their purchases, like getaways and spas. The last kind was special offers and privileges that could not be purchased. These could be talks about interesting subjects open to a selected few, or tickets to a sneak movie preview that was not shown elsewhere.

In conclusion, the findings provide some interesting managerial implications related to the types of rewards offered, as well as the design of incentive programmes. Rewards relevant to the products and services purchased, as well as those given to both the WOM giver and recipient, are relatively more effective in encouraging WOM behaviour and referrals.

 

THE EFFECTS OF SINGLE-MESSAGE SINGLE-SOURCE MIXED WORD-OF-MOUTH INFORMATION ON PRODUCT ATTITUDE AND PURCHASE INTENTION

Ho Lai Ying, Nanyang Technological University, Singapore

Cindy M.Y. Chung, Nanyang Technological University, Singapore

When seeking opinions prior to purchases, it is not uncommon for consumers to hear positive and negative comments about the product. Similarly, consumers sometimes provide positive and negative opinions when asked for their views. Existing word-of-mouth (WOM) literature has documented consumers’ generation of WOM (Richins 1983) and reliance on WOM (Reingen and Kernan 1986), and has shown that WOM information often has a major influence on consumer decision-making (Bayus 1995). However, previous studies mainly investigated the effects of positive WOM (PWOM) and negative WOM (NWOM) separately, presenting subjects with either PWOM or NWOM (Mizerski 1982; Smith and Vogt 1995). Haugtvedt and Wegener (1994) was the only marketing study in which subjects obtained PWOM and NWOM in a single message, but nevertheless from two different sources. Research involving messages with positive and negative information from a single source has mainly been conducted in the psychology domain (Ganzach and Schul 1995; McNeilly & Russ 1989). Therefore, the current research seeks to examine the effects of such messages on the WOM recipient’s product attitude and purchase intention. The findings are expected to add to marketing researchers’ and practitioners’ understanding of this vital source of product and consumption information.

Existing research shows that subjects gave favourable product evaluation when given mainly PWOM, and unfavourable evaluation when given mainly NWOM (Herr, Kardes, and Kim 1991; Wilson and Peterson 1989). Haugtvedt and Wegener (1994) found that roduct evaluations were dependent upon subject’s motivation and ability (i.e., involvement) to process information that contained both positive and negative information. In their study, high-involvement subjects gave greater weight to earlier than later information. As such, a primacy effect occurred whereby a strong position developed after reading the earlier information, leading to a resistance to later information. Low-involvement subjects, on the other hand, were more influenced by the later than earlier information. A recency effect occurred here because a weak position was initially formed, leaving room for later information to overwrite it. However, given that the earlier and later information came from different sources, and given the lack of counterbalancing, this research might have confounded the information presentation order effect with the credibility of the two sources.

TABLE 1

EXPERIMENTAL CONDITIONS IN STUDY 1

TABLE 2

CELL MEANS FOR ATTITUDE & PURCHASE INTENTION IN STUDY 1 AND STUDY 2

The current research attempts to separate out the information presentation order effect from source credibility by presenting information using only one source. Its major prediction is that subjects will give greater weight to earlier (later) information under high (low) involvement.

Study 1 used a one-factor between-subjects design to initially test for an effect of information presentation order on product attitude and purchase intention. Subjects were randomly assigned to one of the six conditions (Table 1), and were presented with information and evaluation on a restaurant by attributes (e.g., service quality, price) as provided by a friend. Subjects imagined having just met up with this friend, and were deciding on a dining place on the spot. Hence, the experimental context was of fairly low involvement.

Two separate one-way ANOVAs showed a presentation order effect for both product attitude (F(5,331)=53.53, p<.001), and purchase intention (F(5,331)=39.82, p<.001). Further cell comparisons showed that subjects in the "P" condition had more favourable attitude and higher purchase intention than those in the "N" condition (p’s<.001), replicating previous findings (Table 2). Most of the cell means for the mixed valence conditions were significantly different from "P" and "N" for both dependent variables. In addition, "NP_R" subjects showed marginally higher purchase intention than subjects in "PN_R" (p=.10). There is initial evidence that presenting PWOM later versus earlier could result in better purchase intention.

Study 2 used a more complete design to measure product attitude and purchase intention. Subjects were randomly assigned to one of the conditions in a 2 (involvement: high vs. low) by 4 (presentation order: P vs. N vs. NP vs. PN) between-subjects design. Two different scenarios in which subjects rated a restaurant were used to manipulate low (same as in Study 1) and high (in preparation of celebrating grandma’s 75^th birthday) involvement. Two separate two-way ANOVAs showed a presentation order main effect for product attitude (F(3,194)=214.04, p<.001), and purchase intention (F(3,197)=106.22, p<.001). Subjects in "P" had more favorable attitude and higher purchase intention than subjects in "N" (p’s<.001) (Table 2). Cell means for the mixed valence conditions for both dependent variables were significantly different from "P" and "N" (p’s<.001). Also, subjects in "NP" had more positive attitude and greater purchase intention than subjects in "PN" (p’s<.001). There was no involvement main effect, or involvement by presentation order interaction effect.

In summary, Study 1 presented subjects with both PWOM and NWOM in a single message from the same source, and obtained initial evidence on a recency effect in purchase intention. Data from Study 2 showed that product attitude and purchase intention were subject to a recency effect regardless of involvement. Results obtained therefore differ from previous findings, indicative of a need for further inquiry.

REFERENCES

Arndt, Johan (1967), Word of Mouth Advertising. New York: Advertising Research Foundation Inc.

Baumeister, Roy F. and Edward E. Jones (1978), "When Self-Presentation is Constrained by the Target’s Knowledge: Consistency and Compensation," Journal of Personality and Social Psychology, 36(6), 608-618.

Bayus, B.L. (1985), "Word of Mouth: the Indirect Effects of Marketing Efforts," Journal of Advertising Research, 25(3), 31-39.

Belk, Russell W. (1988), "Possessions and the Extended Self," Journal of Consumer Research, 15(September), 139-168.

Chaiken, Shelly, Roger Giner-Sorolla, and Serena Chen (1996), "Beyond AccuracyBDefense and Impression Motives in Heuristic and Systematic Information Processing," in The Psychology of ActionBLinking Cognition and Motivation to Behavior, eds. P. M. Gollwitzer and J. A. Bargh, New York: Guildford Press, 553-578.

Day, George S. (1971), "Attitude Change, Media and Word of Mouth," Journal of Advertising Research, 11(6), 31-40.

Dichter, E. (1966), "How Word-of-Mouth Advertising Works," Harvard Business Review, 16, 147-166.

Ganzach, Y. and Y. Schul (1995), "The Influence of Quantity of Information and Goal Framing on Decision," Acta Psychologica, 89, 23-36.

Gitelson, Richard and John Crompton (1983), "The Planning Horizons and Sources of Information used by Pleasure Vacationers," Journal of Travel Research, 21(3), 2-7.

Globe and Mail (1999), "Word of Mouth Seen as Key Marketing Tool," May 12, M1.

Haugtvedt, C.P. and D.T. Wegener (1994), "Message Order Effects in Persuasion: An Attitude Strength Perspective," Journal of Consumer Research, 21(June), 205-208.

Herr, P.M., F.R. Kardes, and J. Kim (1991), "Effects of Word-of-Mouth and Product-Attribute Information on Persuasion: an Accessibility-Diagnosticity Perspective," Journal of Consumer Research, 17(March), 454-462.

Kunda, Ziva (1987), "Motivation and Inference: Self-Serving Generation and Evaluation of Evidence," Journal of Personality and Social Psychology, 53(4), 636-647.

McNeilly, K.M., and F.A. Russ (1989), "Errors in Recall and Judgment: the Effect of Information Availability," Journal of General Psychology, 116, 285-295.

Mizerski, R.W. (1982), "An Attribution Explanation of the Disproportionate Influence of Unfavourable Information," Journal of Consumer Research, 9(December), 301-310

Petty, Richard E. and John T. Cacioppo (1986), "The Elaboration Likelihood Model of Persuasion," in Advances in Experimental Social Psychology, Vol. 19, ed. L. Berkowitz. New York: Academic Press, 123-205.

Reingen, P.H. and J.B. Kernan (1986), "Analysis of Referral Networks in Marketing: Methods and Illustration," Journal of Marketing Research, 23(November), 370-378.

Richins, M.L. (1983), "Negative Word-of-Mouth by Dissatisfied Consumers: a Pilot Study," Journal of Marketing, 47(December), 68-78.

Richins, M.L. (1994), "Special Possessions and the Expression of Material Values," Journal of Consumer Research, 21(December), 522-533.

Rogers, Everett M. (1983), Diffusion of Innovations. New York: Free Press.

Singh, Jagdip (1988), "Consumer Complaint Intentions and Behavior: A Review and Prospect," Journal of Marketing, 52(January), 93-107.

Smith, Robert E. and Tiffany S. Meyer (1980), "Attorney Advertising: A Consumer Perspective," Journal of Marketing, 44(Spring), 56-64.

Smith, Robert E. and C.A. Vogt (1995, "The Effects of Integrating Advertising and Negative Word-of-Mouth Communications on Message Processing and Response," Journal of Consumer Psychology, 4, 133-151.

Swan, John E. and Richard L. Oliver (1989), "Postpurchase Communications by Consumers," Journal of Retailing, 65(4), 516-533.

Westbrook, Robert A. (1987), "Product/Consumption-Based Affective Responses and Postpurchase Processes," Journal of Marketing Research, 24(August), 258-270.

Wilson, W.R. and R.A. Peterson (1989), "Some Limits on the Potency of Word-of-Mouth Information," Advances in Consumer Research, 16, 23-29.

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