Decision Support Systems xxx (2012) xxx–xxx

DECSUP-12005; No of Pages 8

Contents lists available at SciVerse ScienceDirect

Decision Support Systems

j ourna l homepage: www.e lsev ie r .com/ locate /dss

What drives consumers to spread electronic word of mouth in onlineconsumer-opinion platforms

Christy M.K. Cheung a,⁎, Matthew K.O. Lee b,1

a Department of Finance and Decision Sciences, Hong Kong Baptist University, Hong Kongb Department of Information Systems, City University of Hong Kong, Hong Kong

⁎ Corresponding author. Tel.: +852 34112102; fax: +E-mail addresses: ccheung@hkbu.edu.hk (C.M.K. Che

(M.K.O. Lee).1 Tel: +852 27887348; fax: +852 27888694.

0167-9236/$ – see front matter © 2012 Elsevier B.V. Alldoi:10.1016/j.dss.2012.01.015

Please cite this article as: C.M.K. Cheung, M.platforms, Decis. Support Syst. (2012), doi:

a b s t r a c t

a r t i c l e i n f o

Article history:Received 10 August 2010Received in revised form 23 November 2011Accepted 23 January 2012Available online xxxx

Keywords:Electronic word-of-mouth (eWOM)communicationPublic goodOnline consumer reviewsReputationSense of belongingEnjoyment of helpingElectronic marketing

The advance of the Internet facilitates consumers to share and exchange consumption-related advice throughonline consumer reviews. This relatively new form of word-of-mouth communication, electronic word-of-mouth (eWOM) communication, has only recently received significant managerial and academic attention.Many academic studies have looked at the effectiveness of positive eWOM communication, examining theprocess by which eWOM influences consumer purchasing decisions. eWOM behavior is primarily explainedfrom the individual rational perspective that emphasizes a cost and benefit analysis. However, we felt therewas a need for an extensive study that examines consumers' motives for eWOM. In this paper, we focus onthe factors that drive consumers to spread positive eWOM in online consumer-opinion platforms. Buildingon the social psychology literature, we identified a number of key motives of consumers' eWOM intentionand developed an associated model. We empirically tested the research model with a sample of 203 membersof a consumer review community, OpenRice.com. The model explains 69% of the variance, with reputation,sense of belonging and enjoyment of helping other consumers significantly related to consumers' eWOMintention. The results of this study provide important implications for research and practice.

© 2012 Elsevier B.V. All rights reserved.

1. Introduction

With the advent of Internet technologies, traditional word-of-mouth communication has been extended to electronic media, such asonline discussion forums, electronic bulletin board systems, news-groups, blogs, review sites, and social networking sites [34,44]. Every-one can share their opinion and experience related to products withcomplete strangers who are socially and geographically dispersed[19]. This new form of word of mouth, known as electronic word ofmouth (eWOM), has become an important factor in shaping consumerpurchase behavior. Hennig-Thurau et al. [27] argued that informationprovided on consumer opinion sites is more influential amongconsumers nowadays. Industrial statistics have also provided evidencein supporting the significant impact of eWOM communication. Forinstance, eMarketer revealed that 61% of consumers consulted onlinereviews, blogs and other kinds of online customer feedback beforepurchasing a new product or service [22]. In addition, 80% of thosewho plan to make a purchase online will seek out online consumerreviews before making their purchase decision [29]. Some consumers

852 34115585.ung), ismatlee@cityu.edu.hk

rights reserved.

K.O. Lee, What drives consum10.1016/j.dss.2012.01.015

even reported that they are willing to pay at least 20%more for servicesreceiving an “Excellent”, or 5-star, rating than for the same servicereceiving a “Good”, or 4-star rating [15].

Electronic word-of-mouth (eWOM) communication has onlyrecently received significant managerial and academic attention. Mostacademic studies have looked at the effectiveness of eWOM communi-cation, examining the process by which eWOM influences consumerpurchasing decisions. To date, the issue of consumers' eWOM intentionhas received limited attention in the IS literature. We still do notfully understand why consumers spread positive eWOM in onlineconsumer-opinion platforms. Among the few existing publications,eWOM behavior is primarily explained from individual rationalperspectivewith the emphasis on cost and benefit. Consumer participa-tion in online consumer-opinion platforms depends a lot on interac-tions with other consumers. We believe that it is necessary to furtherextend existing work by adopting a diverse theoretical perspective toexplain this new social phenomenon focusing on antecedents toeWOM intentions. In the second section of this paper, we address thetheoretical background. Then, we present our research model andhypotheses; and describe a survey study of users in an onlineconsumer-opinion platform to empirically test the research model.Next, we discuss the findings of our empirical study. And finally, weconclude by describing the implications for both research and practice,the limitations of the study, and future research directions.

ers to spread electronic word of mouth in online consumer-opinion

2 C.M.K. Cheung, M.K.O. Lee / Decision Support Systems xxx (2012) xxx–xxx

2. Theoretical background

Prior literature provides a rich foundation of theory on which tobuild a research model that explains why consumers are willing tospread positive eWOM in online consumer-opinion platforms. In thissection, we first define electronic word-of-mouth communication andcompare the concept with traditional word-of-mouth communication.We then describe the theoretical foundation of our research model.

2.1. Definition of eWOM communication

With the advent of the Internet, there has been a paradigm shift inword-of-mouth communication. Traditional word-of-mouth (WOM),which was originally defined as an oral form of interpersonal non-commercial communication among acquaintances [5], has evolvedinto a new form of communication, namely electronic word-of-mouth(eWOM) communication. eWOM communication can take place invarious settings. Consumers can post their opinions, comments andreviews of products on weblogs (e.g. xanga.com), discussion forums(e.g. zapak.com), review websites (e.g. Epinions.com), retail websites(e.g., Amazon.com), e-bulletin board systems, newsgroup and socialnetworking sites (e.g. facebook.com).

eWOM differs from traditional WOM in many ways. First, unliketraditional WOM, eWOM communications possess unprecedentedscalability and speed of diffusion. eWOM communications involvemulti-way exchanges of information in asynchronous mode [27,28].The use of various electronic technologies such as online discussionforums, electronic bulletin boards, newsgroups, blogs, review sitesand social networking sites facilitate information exchange amongcommunicators [33]. Second, eWOM communications are more per-sistent and accessible than traditional WOM. Most of the text-basedinformation presented on the Internet is archived and thus, in manycases, at least in theory, is available for an indefinite period of time[28,37]. Third, eWOM communications are more measurable thantraditional WOM. The presentation format, quantity and persistenceof eWOM communications have made them more observable. Lastly,the electronic nature of eWOM in most applications may dampen thereceiver's ability to judge the sender and his or her message on factorssuch as credibility. People can only judge the credibility of the com-municator based on the associated cues through online reputationsystems (online ratings, website credibility, etc.).

2.2. Prior research on eWOM communication

The topic of electronic word-of-mouth (eWOM) communicationis generating increased interest in business disciplines such as mar-keting, consumer behavior, economics, and information systems. Re-searchers have adopted various research approaches to examine thisimportant phenomenon. Most of these different research approachesexplore the impact of eWOM communication. However, research onwhy consumers engage in eWOM in online consumer-opinion plat-forms remains relatively limited. A prominent study of eWOMcommunication motives is by Henning-Thurau et al. [27]. They builton Balasubramanian and Mahajan [6], identifying five main motiva-tional categories of positive eWOM communication: focus-relatedutility (concern for other consumers, helping the company, socialbenefits, and exerting power), consumption utility (post-purchaseadvice-seeking), approval utility (self-enhancement and economic re-wards), moderator-related utility (convenience and problem-solvingsupport), and homeostase utility (expressing positive emotions andventing negative feelings). Sun et al. [39] also proposed an integratedmodel to explore the antecedents and consequences of eWOM in thecontext of music-related communication. They found that innovative-ness, internet usage, and internet social connection are significantfactors in eWOM behavior. Tong et al. [41] explored costs (cognitivecost and executional cost) and benefits (enjoyment in helping other

Please cite this article as: C.M.K. Cheung, M.K.O. Lee, What drives consumplatforms, Decis. Support Syst. (2012), doi:10.1016/j.dss.2012.01.015

consumers and enjoyment in influencing the company, self-enhancement, and economic reward) of consumer's information contri-bution to online feedback systems. These studies provide a reasonablestart to exploring further the motives behind eWOM communicationin away that does not necessarily approach eWOMbehavior as individ-ual rational phenomenon.

2.3. The public good

In the literature, information sharing is viewed as a public-good phe-nomenon. A public good is characterized as “a shared resource fromwhich every member of a group may benefit, regardless of whether ornot they personally contribute to its provision, and whose availabilitydoes not diminish with use” (p. 693) [11]. The fundamental problemof a public good is that any individual may consume a public goodwith-out contributing to a group. This results in a social dilemma situation,which occurs when an individual attempts to maximize self-interestover social-interest and makes a rational decision. In the online envi-ronment, anyone can access and consume knowledge without makinga direct contribution back to it. It is very likely that individuals willfree-ride [9,30]. Wasko and Tiegland [43] however urged that thoughpublic goods are subjected to social dilemmas, they are nonethelesscreated and maintained through collective action. In other words,public goods are still shared and contributed to voluntarily throughcooperation of individuals. Based on the social psychology literature,we identified four perspectives that explain why consumers spreadeWOM in online consumer-opinion platforms: egoism, collectivism,altruism, and principlism.

Egoism refers to serving the public good to benefit oneself. Re-searchers in psychology, sociology, economics, and political sciencesassume that all human actions are ultimately directed toward self-interest. Rewards and avoidance are the most obvious self-benefitsthat drive individuals to act for the public good. Collectivism refers toserving the public good to benefit a group. The act for the public goodis for the group's benefit, as the self shifts frompersonal self to collectiveself. This is the most widely accepted social psychology theory of groupbehavior. Altruism refers to serving the public good to benefit one ormore others. The motive for the public good can be linked to empathicemotion. Empathy (feelings of sympathy, compassion, tenderness,and the like) is a source of altruism. Some researchers have shownthat feeling empathy for a person in need leads to increased helpingof that person [20]. Principlism refers to serving the public good touphold a principle. The motivation is to uphold, typically, some moralprinciple, such as justice or the utilitarian principle of the greatestgood for the greatest number. Gorsuch and Orberg [24] found that inmoral situations, people reported their intentions to act out of theirsense of moral responsibility.

2.4. Knowledge self-efficacy

Prior studies [33] have demonstrated that knowledge self-efficacy isan important antecedent of knowledge sharing in the online environ-ment. Individuals tend to provide useful advice on computer networksif they possess a high level of expertise [17]. Conversely, when theylack information or knowledge which is useful to others, they tend tomake less contribution in knowledge sharing since, for example theybelieve that they cannot make a positive impact for the organization[30]. Insufficient knowledge self-efficacy also hinders individuals toshare in web-based discussion boards [33].

This line of study suggests that people form beliefs about whatthey can do, predict likely outcomes of prospective actions, and setgoals for themselves in order to achieve desired outcomes. In otherwords, the motivations of performing a behavior do not stem fromthe goals themselves, but from the self-evaluation that is made condi-tional on their fulfillment. Bandura [7] defines perceived self-efficacyas “people's beliefs about their capabilities to produce designated

ers to spread electronic word of mouth in online consumer-opinion

3C.M.K. Cheung, M.K.O. Lee / Decision Support Systems xxx (2012) xxx–xxx

levels of performance that exercise influence over events that affecttheir lives” (p. 71). Self-efficacy is created through mastery experi-ence. Success builds a strong belief in one's self-efficacy and moti-vates an individual to continue the behaviors.

3. Research model and hypotheses

Based on the literature reviewed above, we develop a model of an-tecedents to eWOM intentions in online consumer-opinion platforms,depicted in Fig. 1. The antecedent variables are from four differenttheoretical perspectives as well as knowledge self-efficacy. Ourfocus is on intentions to behave as, however indeed a relationship be-tween intention and behavior is well established [1]. In this section,the key components of the research model and their interrelation-ships are addressed.

3.1. Egoistic motivation

Amotive is considered egoistic if the ultimate goal is to increase theactor's own welfare [8]. Individuals are deemed as egoistic when theyaim at tangible or intangible returns after sharing information withothers. Social exchange theory has been adopted to explain the actionfor the public good in terms of egoism in recent years [9,30]. Being ratio-nal, human beings try to look for returns (e.g. pay, prizes, reputation,and recognition) by maximizing their benefits and minimizing theircost during information exchange process with others [32].

This perspective has been widely adopted in many eWOM commu-nication publications [27,41]. For example, reputation is often cited asan important determinant of information sharing behavior [16,17]. Peo-ple share and contribute their knowledge because they want to gain aninformal recognition and establish themselves as experts [43]. Similarly,we believe that if a consumer wants to gain a reputation in an onlineconsumer-opinion platform, he/she has a higher tendency to spreadeWOM. This leads to the following hypothesis:

H1. The perception of the opportunity to enhance one's own reputa-tions is positively related to one's eWOM intention.

Fig. 1. Researc

Please cite this article as: C.M.K. Cheung, M.K.O. Lee, What drives consumplatforms, Decis. Support Syst. (2012), doi:10.1016/j.dss.2012.01.015

Another egoistic motivator of the act for the public good is reci-procity, which is also conceived as a benefit for individuals to engagein social exchange. When information providers do not know eachother, the kind of reciprocity that is relevant is called “generalized”exchange [21], and the person who offers help to others is expectingreturns in the future [32]. Prior research found that people who shareknowledge in online communities value reciprocity [42], and it is thisbelief that drives them to participate and share. Thus, this leads to thefollowing hypothesis:

H2. The perception of the opportunity for reciprocity is positively re-lated to one's eWOM intention.

3.2. Collective motivation

Collectivism is defined as the motivation with the ultimate goal ofincreasing the welfare of a group or collective [8]. In other words, in-dividuals with a collective motive contribute their knowledge for thebenefit of the whole group rather than personal return. In terms of ac-tion for the public good, collectivism can be linked to social identitytheory, in which individuals gain social identity from the groupsthey belong to [40]. When individuals identify themselves as mem-bers of a social aggregate, they are more likely to define themselvesin terms of their membership in that group [18]. Members have thefeeling that others' needs will be satisfied by the resources receivedthrough their contributions to the group [35].

Sense of belonging refers to a sense of emotional involvementwith the group. When people identify themselves as part of the com-munity and align their goals with those of the community, they willtreat other members as their kin and they will be willing to do some-thing beneficial to/for others that are not necessarily beneficial [26].Lakhani and Von Hipper [32] also argued that committed electronicnetwork members take part in knowledge sharing since they thinksuch behavior is best for the community. Hence, people with this var-iant of intrinsic motivation will be motivated to participate in sharingactivities and help their kinship partners.

h model.

ers to spread electronic word of mouth in online consumer-opinion

4 C.M.K. Cheung, M.K.O. Lee / Decision Support Systems xxx (2012) xxx–xxx

H3. The opportunity for the sense of belonging is positively related toone's eWOM intention.

3.3. Altruistic motivation

Altruism is motivation with the ultimate goal of increasing thewelfare of one or more individuals other than oneself [8]. Individualsacting on altruistic goals are willing to volunteer themselves to con-tribute their knowledge to online consumer reviews without expect-ing direct rewards in return. For example, consumers may sharepurchasing experience just because others have a need for it [31].When studied in terms of empathic emotion, individuals may haveempathy toward a person in need and this increases helping of thatperson [8].

Enjoyment of helping has been acknowledged by researchers as analtruistic factor to explain individuals' willingness to share knowledgein electronic networks of practice or online social spaces [27,30,41].Though there is no apparent compensation, people in virtual communi-ties still obtain intrinsic enjoyment and satisfaction by helping othersthrough sharing their knowledge [4,31,42]. Hence,

H4. The opportunity to realize personal enjoyment is positively relatedto one's eWOM intention.

3.4. Principlistic motivation

Principlism refers to themotivation towards the ultimate goal of up-holding somemoral principle, such as justice or the utilitarian principleof the greatest good for the greatest number [8]. The predictive power ofprinciplistic motivation on behavioral intention has been supported byvarious empirical studies [24]. Action for the public good in terms ofprinciplism can be explained by normative commitment, inwhich com-mitment is a sense of obligation to the organization [3,36].With a strongsense of commitment to the community, individuals in virtual commu-nities are more likely to feel obliged to help others by contributingknowledge [18]. They are willing to contribute their knowledge to thewell being of the organization [36].

Moral obligation is derived from principlism. Commitment to onlinecommunities conveys a sense of duty or obligation to help others on thebasis of shared membership [41]. In the context of an organization,people view their knowledge as a public good and they are motivatedto have knowledge exchange with others because of moral obligationand community interest [3]. In online communities, individuals with astrong sense of commitment to the community are more likely to feelobliged to help others by contributing knowledge [43]. Therefore, webelieve that when a consumer has a strong sense of moral obligation,there will be a higher chance for them to spread eWOM in onlineconsumer-opinion platforms.

H5. The opportunity to feel a moral obligation is positively related toone's eWOM intention.

3.5. Knowledge self-efficacy

In social cognitive theory, self-efficacy is a personal judgment ofone's capability to execute actions required for designated types ofperformances. It has a great impact on people's intentions and behavior[7]. Derived from this line of study, knowledge self-efficacy can beserved as a self-motivator for knowledge contribution in onlineplatforms. Previous studies have already illustrated the importance ofknowledge self-efficacy on people's intention to share knowledge[30]. We also believe that a higher knowledge self-efficacy about apurchasing experience, leads to a higher tendency to spread eWOM inonline consumer-opinion platforms.

Please cite this article as: C.M.K. Cheung, M.K.O. Lee, What drives consumplatforms, Decis. Support Syst. (2012), doi:10.1016/j.dss.2012.01.015

H6. The degree of perceived knowledge self-efficacy is positively re-lated to one's eWOM intention.

4. Research method

The research model was examined using a sample of onlineconsumer-opinion platform users from OpenRice.com. OpenRice.com,one of the most successful online communities in Hong Kong, sharesinformation about 15,000 restaurants in Hong Kong and Macau. It is agood search tool with all restaurant information categorized in termsof the style of food, location of the restaurant, price ranges, and thelike.

4.1. Data collection

In this study, the sample frame was individuals who have usedOpenRice.com. A convenience sample was used by inviting volunteersto participate in this study. We posted an invitation message with theURL to the online questionnaire on a number of Facebook groups re-lated to dining experiences in Hong Kong. To increase the responserate, entry in a lottery for supermarket vouchers was offered as an in-centive for participation.

4.2. Sample profile

The respondents were asked to complete the questionnaire basedon their experience with OpenRice.com. A total of 203 usable ques-tionnaires were collected in this study. Among the 203 respondents,57% was female and 43% was male. A majority of our respondents(67%) were aged between 21 and 25. 78% of our respondents hadan education level of university or above.

4.3. Measures

The constructs of interest in this study included consumers' eWOMintention, reputation, reciprocity, sense of belonging, enjoyment ofhelping, moral obligation, and knowledge self-efficacy. We used estab-lished measures from previous literature (See Appendix A). All con-structs were measured using multi-item perceptual scales and werecarried out by a seven-point Likert scale, ranging from strongly disagree(1) to strongly disagree (7).

5. Data analysis and results

The Partial Least Squares (PLS) method was used to perform thestatistical analysis in this study. PLS technique provides a better ex-planation for complex relationships [23] and is widely adopted by ISresearchers [13]. Moreover, it is suitable when the focus of the re-search is on theory development. Following the two-step analyticalapproach [25], we first conducted the psychometric assessment ofour measurement scales, and we then evaluated the structuralmodel. Using this approach, we have a higher confidence that theconclusion on structural relationship is drawn from a set of measure-ment instruments with desirable psychometric properties.

5.1. Measurement model

The convergent validity and discriminant validity of the constructsin our model were examined. Convergent validity was tested usingthree criteria of all constructs: (1) the composite reliability (CR)should be at least 0.70 [13], (2) the average variance extracted(AVE) should be at least 0.50 [23], and (3) all item loadings shouldbe greater than 0.707 [13]. Results of our analysis are shown inTable 1. All three conditions of convergent validity were satisfied inour data sample by having the CRs ranging from 0.89 to 0.96, and

ers to spread electronic word of mouth in online consumer-opinion

Table 1Psychometric properties of measures.

Construct Item Loading t-value Mean St. dev

Knowledge self-efficacyCR=0.91; AVE=0.84

SE1 0.88 14.64 4.91 1.38SE2 0.95 56.50 4.27 1.46

Enjoyment of helpingCR=0.96, AVE=0.89

EH1 0.93 78.81 4.15 1.48EH2 0.95 83.73 4.28 1.45EH3 0.95 111.36 4.31 1.40

Consumers' eWOM intentionsCR=0.92, AVE=0.79

INT1 0.90 57.23 3.94 1.55INT2 0.91 61.07 3.70 1.57INT3 0.85 32.61 4.03 1.51

Moral obligationCR=0.92, AVE=0.79

MO1 0.93 96.38 3.98 1.46MO2 0.86 28.77 4.22 1.54MO3 0.88 43.16 3.59 1.48

ReputationCR=0.96, AVE=0.93

RP1 0.96 179.09 3.58 1.51PR2 0.97 190.60 3.50 1.54

Sense of belongingCR=0.94, AVE=0.77

SB1 0.87 44.04 3.81 1.59SB2 0.81 23.14 4.07 1.49SB3 0.91 72.30 3.68 1.52SB4 0.90 62.31 3.63 1.48SB5 0.88 63.59 3.52 1.60

ReciprocityCR=0.89; AVE=0.67

RC1 0.77 5.41 4.61 1.28RC2 0.81 5.04 5.00 1.38RC3 0.88 4.43 4.98 1.32RC4 0.80 3.53 4.86 1.36

Notes: CR—Composite Reliability, AVE—Average Variance Extracted.

5C.M.K. Cheung, M.K.O. Lee / Decision Support Systems xxx (2012) xxx–xxx

the AVEs from 0.67 to 0.93. The item loadings were all higher than the0.707 benchmark.

Discriminant validity is indicated by low correlations between themeasure of interest and the measure of other constructs [23]. This va-lidity can be assessed by having the square root of the average vari-ance extracted (AVE) of each construct higher than the correlationsbetween it and all other constructs. As shown in Table 2, the squareroot of the AVE of each construct is located on the diagonal of thetable and is in bold. A reasonable degree of discriminant validity ob-tains since each of them is greater than the correlations between itand all other constructs. We followed Segars and Grover's [38] guide-line and further tested the correlations between enjoyment of help-ing, moral obligation, and sense of belonging. First, a modelimposing a correlation of 1 between the two specific constructs isrun. Then, another model with a freely estimated correlation betweenthe two constructs is run. Discriminant validity is demonstrated ifthere is a significant difference of the Chi-square statistics (i.e., Chi-square difference is greater than 3.84) between the constrained (thecorrelation between constructs is set) and unconstrained models(the correlation between constructs is free). In the current study, ro-bust evidence of convergent validity and discriminant validity wasfound with these data.

5.2. Structural model

The structural model analysis was assessed based on the test ofthe hypothesized effects in our research model. Fig. 2 shows theresults of the hypothesized structural model test, including the vari-ance explained (R2 value) of the dependent variable, estimated path

Table 2Correlation matrix and psychometric properties of key constructs.

EH INT MO RC RP SE SB

Enjoyment of Helping (EH) 0.94Consumers' eWOM Intentions (INT) 0.73 0.89Moral Obligation (MO) 0.71 0.68 0.89Reciprocity (RC) 0.20 0.13 0.09 0.82Reputation (RP) 0.52 0.63 0.66 0.09 0.96Knowledge Self-Efficacy (SE) 0.17 0.22 0.11 0.59 0.12 0.92Sense of Belonging (SB) 0.72 0.79 0.75 0.16 0.73 0.22 0.87

Notes: Italicized diagonal elements are the square root of AVE for each construct. Off-diagonal elements are the correlations between constructs.

Please cite this article as: C.M.K. Cheung, M.K.O. Lee, What drives consumplatforms, Decis. Support Syst. (2012), doi:10.1016/j.dss.2012.01.015

coefficients with significant paths indicated by asterisks, and associatedt-values of the paths. Bootstrap resampling procedure was used to per-form the significant testing for each path.

An examination of the R2 value demonstrates that the modelexplains a substantial amount of the variance in the outcome variable.In our model, it explains 69% of the variance in consumers' eWOMintention. The significant antecedents are reputation, sense of belongingand enjoyment of helping, with path coefficients at 0.11, 0.41 and 0.33respectively. This provides support for H1, H3 and H4.

6. Discussion and conclusion

Given the limited research in the area of consumers' intention tospread eWOM in online consumer-opinion platforms, this study seeksto consider the factors that shape eWOM behavior. This section dis-cusses the results of hypothesis testing of the researchmodel, addressesthe limitations of the study, and highlights the contributions to researchand practice.

6.1. General discussion

The research model gains much of its theoretical foundation fromthe social psychology literature. The analysis shows that consumers'eWOM intention is significantly related to three antecedents, reputa-tion (marginal significance), sense of belonging, and enjoyment ofhelping.

Sense of belonging had relatively the most impact on consumers'eWOM intention. The result is consistent with previous eWOMmarketing literature, where affective commitment (sense of belong-ing) is an essential ingredient that fosters loyalty and citizenship ina group [18]. In our case, consumers who have a stronger sense ofbelonging to OpenRice.com have greater citizenship intentions(e.g., sharing dining experiences with other consumers). This alsoillustrated the importance of including social factors in the currentinvestigation. Our study also showed that enjoyment of helping othersis crucial in affecting consumers' eWOM intention. Intentions to writeabout dining experiences in OpenRice.com demonstrate enjoyment ofhelping others. Consumers can benefit other community membersthrough helping them with their purchasing decisions. Specifically,this act can save others from having negative experiences when visitingsubstandard restaurants. Reputation is a marginally significant factoraffecting consumers' eWOM intention. Consumers spreading eWOMin online consumer-opinion platforms related to a desire to alter repu-tation. These consumer-opinion platforms have enormous potential forscale and reach. Some consumers are willing to contribute dining expe-riences because they may want to be viewed as an expert by a largegroup of consumers.

Reciprocity, moral obligation and knowledge self-efficacy did notdemonstrate a significant relationship with consumers' eWOM inten-tion. Unlike internal knowledge sharing systems, members on OpenRi-ce.com post their reviews based on their experiences in visiting specificrestaurants. The opinion in these online reviews helps other diners tojudge whether the restaurants are worth visiting. The experience theyshare does not necessary lead to a future request for knowledge beingmet. The results are consistent with some research which shows reci-procity does not influence the intention to use a knowledgemechanism[14]. In addition, as OpenRice.com is an informal consumer-basedcommunity, members may have sense of belonging, but the commit-ment to OpenRice.com does not necessarily convey a sense of duty orobligation to help others on the basis of shared membership. Providingconsumer reviews is on voluntary basis, which means users have theright to decide if they would like to leave their comments. Principlismmight have more impact when the obligation is stipulated in explicitterms. For instance, the moderator of the platform should include theterms of use (e.g., with an emphasis on the obligation to share andhelp other users) during user registration. Finally, whether they are

ers to spread electronic word of mouth in online consumer-opinion

(Note: *p<0.10, **p<0.05, ***p<0.01)

Fig. 2. Result of the research model.

6 C.M.K. Cheung, M.K.O. Lee / Decision Support Systems xxx (2012) xxx–xxx

frequent patrons or first-time diners, all are welcome to provide re-views about restaurants they have visited. Reviewers in OpenRice.commay bear no thorough understanding about the restaurants and maysimply express their opinion based on the service quality they received.Thus, knowledge self-efficacy does not have a significant impact onconsumers' eWOM intentions in online consumer-opinion platforms.

6.2. Limitations and future research directions

In interpreting the results of this study, one must pay attention to anumber of limitations. Our review of prior literature indicates thatresearch on consumer engagement in eWOM communication remainsrelatively newandhas only received limited attention in the scholarly lit-erature. To enhance the understanding of this phenomenon andcontribute towards the developing of the existing literature in this area,we propose a theoretical model that explains consumers' eWOM inten-tion. In the current investigation, we included only the key motivesfrom each of the four perspectives of the social psychology literature.Though the explanatory power of our researchmodel is high, we believethat future research studies should include someother related constructs(e.g., rewards, subjective norm, costs, etc.) to account for the remainingunexplained variance in consumers' eWOM intention. As prior studieshave found that positive eWOM is more likely to occur than negativeeWOM, in the current study, we only focused on consumers' intentionto spread positive eWOM. In line with recent research showing the neg-ativity bias in online consumer behavior [12], future studies should con-tinue to explore the motives that drive users to spread negative eWOM.

The sample size is relatively small and it is a convenience samplecomprisedmostly of students. This suggests that future research shouldinclude amore diverse sample of potential users in different age catego-ries, professions, and usage experience with the consumer-opinionplatforms. A larger sample size can also bring more statistical power

Please cite this article as: C.M.K. Cheung, M.K.O. Lee, What drives consumplatforms, Decis. Support Syst. (2012), doi:10.1016/j.dss.2012.01.015

for analysis. Finally, since only a single questionnaire was used to mea-sure all the constructs in our study, common method bias may exist inthe measurement. Further studies could test our model by using differ-ent research methods to overcome this weakness.

6.3. Implications

Though existing academic research has significantly advanced ourunderstanding of electronic word-of-mouth (eWOM), much of it isfocused on how online consumer reviews affect sales of products andservices. Limited attention has been devoted to the antecedents ofeWOM. In view of this, we attempt to investigate consumers' eWOMintention in the current study. We believe that this study contributesto the conceptual and empirical understanding of eWOM intentions inonline consumer-opinion platforms. Implications of this study are note-worthy for both researchers and practitioners.

This study contributes to existing eWOM research in several ways.First, a lot of existing eWOM studies focus primarily on the impact ofeWOM on consumer purchasing decision. There is a lack of under-standing of how and why consumers are willing to spend their owntime to share their purchasing experiences with other people in theonline environment. This study enriches the existing literature byproposing a theoretical model that explains consumers' eWOM inten-tion. Second, the research model gains its theoretical foundation fromthe social psychology literature and social cognitive theory. Particu-larly, we provided empirical support that social factor such as senseof belonging, also exhibits significant impact on eWOM intentions inonline consumer-opinion platforms. The empirical investigationdemonstrates the relative importance of various antecedent factorsfor consumers' eWOM intention.

The finding of this research is also useful for online consumer-opinion platforms' moderators in understanding their members'

ers to spread electronic word of mouth in online consumer-opinion

Enjoyment of helping (modified from [43])EH1 I like helping other members in OpenRice.com. (Extremely disagree/

Extremely agree)EH2 It feels good to help others other members in OpenRice.com. (Extremely

disagree/Extremely agree)EH3 I enjoy helping other member in OpenRice.com. (Extremely disagree/

Extremely agree)

Moral obligation (modified from [10])MO1 My conscience calls me to contribute and share in OpenRice.com.

(Extremely disagree/Extremely agree)MO2 My decision to share or not in OpenRice.com is fully in line with my moral

conviction. (Extremely disagree/Extremely agree)MO3 I feel morally obliged to share in OpenRice.com. (Extremely disagree/

Extremely agree)

Consumers' eWOM intention (modified from [9])INT1 I intend to share my dining experiences with other members in

OpenRice.com more frequently in the future. (Extremely disagree/Extremely agree)

INT2 I will always provide my dining experiences at the request of othermembers in OpenRice.com. (Extremely disagree/Extremely agree)

INT3 I will try to share my dining experiences with other members inOpenRice.com in a more effective way (Extremely disagree/Extremelyagree)

Knowledge self-efficacy (modified from [30])SE1 I have confidence in my ability to provide knowledge/Information that

others in Open Rice.com consider valuable. (Extremely disagree/Extremelyagree)

SE2 I have the expertise needed to provide valuable knowledge/Information forOpen Rice.com. (Extremely disagree/Extremely agree)

7C.M.K. Cheung, M.K.O. Lee / Decision Support Systems xxx (2012) xxx–xxx

behaviors. The results of this study show that sense of belonging to thecommunity, reputation, and enjoyment of helping others are the mostcritical factors that encourage consumers to share their experienceswith others in the context of online consumer-opinion platforms. Hereare someguidelines for online consumer-opinion platforms'moderators:

▪ Sense of belonging: To enhance consumers' sense of belonging toan online consumer-opinion platform, platform moderatorsshould allow consumers to create their own personal profile. Sim-ilar to social networking platforms such as Facebook, adding otherusers as friends and directly communicating with them may cre-ate a stronger sense of belonging to the group.

▪ Reputation: To encourage more consumers to share their opinions,online consumer-opinion platforms should apply reputation-tracking mechanisms to recognize contributors. Apart from thenumber of contributions, publicly visible cues such as length ofmembership and membership status should be incorporated intothe platform design.

▪ Enjoyment of helping: Online opinion-platforms should provide amechanism where members who have provided useful suggestionsto othermembers are identified and informed that they have helpedothers. Connecting contributors and readers via person-to-personmessaging/chat function can enable readers to show their apprecia-tion for the reviews received.

In conclusion, electronic word-of-mouth communication in onlineconsumer opinion platform represents new and important e-marketing phenomenon, we hope that it triggers additional theorizingand empirical investigation aimed at better understanding of eWOMcommunication in social media.

Acknowledgment

The authors acknowledge with gratitude the generous support ofthe Hong Kong Baptist University for the project (HKBU 240609)without which the timely production of the current report/publica-tion would not have been feasible. In addition, the work describedin this paper was partially supported by a grant from City Universityof Hong Kong (Project No. 7002640).

Appendix A. Measures

Reputation (modified from [42])RP1 I feel that my participation in OpenRice.com improves my status in the

profession. (Extremely disagree/Extremely agree)RP2 I participate in OpenRice.com to improve my reputation in the profession.

(Extremely disagree/Extremely agree)

Reciprocity (RC1-3: modified from [30]; RC4: modified from [43])RC1 When I share my knowledge through OpenRice.com, I believe that I will get

an answer for giving an answer. (Extremely disagree/Extremely agree)RC2 When I share my knowledge through OpenRice.com, I expect somebody to

respond when I'm in need (Extremely disagree/Extremely agree)RC3 When I contribute knowledge to OpenRice.com, I expect to get back

knowledge when I need it (Extremely disagree/Extremely agree)RC4 I know that other members in OpenRice.com will help me, so it's only fair to

help other member. (Extremely disagree/Extremely agree)

Sense of belonging (modified from [2])SB1 I am very attached to OpenRice.com community. (Extremely disagree/

Extremely agree)SB2 Other OpenRice.com members and I share the same objectives. (Extremely

disagree/Extremely agree)SB3 The friendships I have with other OpenRice.com members mean a lot to me.

(Extremely disagree/Extremely agree)SB4 If OpenRice.com members planned something, I would think of as

something “we” would do rather than something “they” would do.(Extremely disagree/Extremely agree)

SB5 I see myself as a part of OpenRice.com. (Extremely disagree/Extremelyagree)

Please cite this article as: C.M.K. Cheung, M.K.O. Lee, What drives consumplatforms, Decis. Support Syst. (2012), doi:10.1016/j.dss.2012.01.015

References

[1] I. Ajzen, From intentions to actions: a theory of planned behavior, in: J. Kuhl, J.Beckmann (Eds.), Action Control: From Cognition to Behavior, Springer-Verlag,Berlin, Heidelberg, New York, 1985.

[2] R. Algesheimer, U.M. Dholakia, A. Hermann, The social influence of brand commu-nity: evidence from European car clubs, Journal of Marketing 69 (3) (2005)19–34.

[3] N.J. Allen, J.P. Meyer, Affective, continuance, and normative commitment to theorganization: an examination of construct validity, Journal of Vocational Behavior49 (3) (1996) 252–276.

[4] R. Arakji, R. Benbunan-Fich, M. Koufaris, Exploring contributions of public resourcesin social bookmarking systems, Decision Support Systems 47 (3) (2009) 245–253.

[5] J. Arndt, Role of product-related conversations in the diffusion of a new product,Journal of Marketing Research (4) (1967) 291–295.

[6] S. Balasubramanian, V. Mahajan, The economic leverage of the virtual communi-ty, International Journal of Marketing Research 5 (3) (2001) 103–138.

[7] A. Bandura, Social Foundations of Thought and Action: a Social Cognitive Theory,Prentice-Hall, Englewood, NJ, 1986.

[8] C.D. Batson, Why act for the public goods? Four answers, Personality and SocialPsychology 20 (5) (1994) 603–610.

[9] G.W. Bock, R.W. Zmud, J.N. Lee, Behavioral intention formation in knowledgesharing: examining the roles of extrinsic motivators, social–psychological forces,and organizational climate, MIS Quarterly 29 (1) (2005) 87–112.

[10] M. Bosnjak, T.L. Tuten, W.W. Wittmann, Unit non response in web access panelsurveys: an extended planned-behavior approach, Psychology and Marketing22 (16) (2005) 489–505.

[11] A. Cabrera, F.E. Cabrera, Knowledge-sharing dilemmas, Organizational Studies 23(5) (2002) 687–710.

[12] C.M.K. Cheung, M.K.O. Lee, User satisfaction with an internet-based portal: anasymmetric and nonlinear approach, Journal of the American Society for Informa-tion Science and Technology 60 (1) (2009) 111–122.

[13] W.W. Chin, in: G. Marcoulides (Ed.), The Partial Least Squares Approach to Struc-tural Equation Modeling, Lawrence Erlbaum Associates, Mahwah, NJ, 1998,pp. 189–217.

[14] C.M. Chiu, M.H. Hsu, E.T.G. Wang, Understanding knowledge sharing in virtualcommunities: an integration of social capital and social cognitive theories, Deci-sion Support Systems 42 (3) (2006) 1872–1888.

[15] ComScore Inc., Online consumer-generated reviews have significant impact onoffline purchase, http://www.comscore.com/Press_Events/Press_Releases/2007/11/Online_Consumer_Reviews_Impact_Offline_Purchasing_Behavior2007.

[16] D. Constant, S. Kiesler, L. Sproull, What's mine is ours, or is it? A study of attitudesabout information sharing, Information Systems Research 5 (4) (1994) 400–421.

[17] D. Constant, L. Sproull, S. Kiester, The kindness of strangers: the usefulness of elec-tronic weak ties for technical advice, Organization Science 7 (2) (1996) 119–135.

[18] U.M. Dholakia, R.P. Bagozzi, L.K. Pearo, A social influence model of consumer par-ticipation in network- and small-group-based virtual communities, InternationalJournal of Research in Marketing 21 (3) (2004) 241–263.

ers to spread electronic word of mouth in online consumer-opinion

8 C.M.K. Cheung, M.K.O. Lee / Decision Support Systems xxx (2012) xxx–xxx

[19] W. Duan, B. Gu, A.B. Whinston, Do online reviews matter? — An empirical inves-tigation of panel data, Decision Support Systems 45 (4) (2008) 1007–1016.

[20] N. Eisenberg, P.A. Miller, The relation of empathy to prosocial and related behaviors,Psychological Bulletin 101 (1) (1987) 91–119.

[21] P. Ekeh, Social Exchange Theory: The Two Traditions, Harvard University Press,Cambridge, MA, 1974.

[22] eMarketer.com., Online review sway shoppers, http://www.emarketer.com/Article.aspx?R=10064042008Last accessed.

[23] C. Fornell, D.F. Larcker, Evaluating structural equation models with unobservablevariables and measurement error, Journal of Marketing Research 18 (1981) 39–50.

[24] R.L. Gorsuch, J. Orberg, Moral obligation and attitudes: their relation to behaviouralintentions, Journal of Personality and Social Psychology 44 (5) (1983) 1025–1028.

[25] J.F. Hair, W.C. Black, B.J. Babin, R.E. Anderson, R.L. Tatham, Multivariate Data Anal-ysis, 6th Eds Pearson Prentice Hall, New Jersey, 2006.

[26] A. Hars, S. Ou, Working for Free? Motivations of participating in open source pro-jects, International Journal of Electronic Commerce 6 (3) (2002) 25–39.

[27] T. Henning-Thurau, K.P. Gwinner, G. Walsh, D.D. Gremler, Electronic word ofmouth via consumer-opinion platforms: what motivates consumers to articulatethemselves on the Internet, Journal of Interactive Marketing 18 (1) (2004) 38–52.

[28] K. Hung, S. Li, The influence of eWOM on virtual consumer communities: socialcapital, consumer learning, and behavioral outcomes, Journal of AdvertisingResearch 47 (4) (2007) 485.

[29] Infogroup Inc., Online consumer reviews significantly impact consumer purchasingdecisions, http://www.opinionresearch.com/fileSave/Online_Feedback_PR_Final_6202008.pdf2009.

[30] A. Kankanhalli, B.C.Y. Tan, K.K. Wei, Contribution knowledge to electronic knowl-edge repositories: an empirical investigation, MIS Quarterly 29 (1) (2005)113–143.

[31] P. Kollock, The economies of online cooperation: gifts, and public goods in cyber-space, in: M.A. Smith, P. Kollock (Eds.), Communities in Cyberspace, Routledge,New York, 1999, pp. 220–239.

[32] K.R. Lakhani, E. Von Hipper, How open source software works: ‘free’ user-to-userassistance, Research Policy 32 (6) (2003) 923–943.

[33] M.K.O. Lee, C.M.K. Cheung, K.H. Lim, C.L. Sia, Understanding customer knowledgesharing in web-based discussion boards: an exploratory study, Internet Research16 (3) (2006) 289–303.

[34] F. Li, T.C. Du, An ontology-based opinion leader identification framework forword-of-mouth marketing in online social blogs, Decision Support Systems 51(1) (2011) 190–197.

[35] D.W. McMillan, D.M. Chavis, Sense of community: a definition and theory, Journalof Community Psychology 14 (1) (1986) 6–23.

[36] R.T. Mowday, R.M. Steers, L. Porter, The measurement of organizational commit-ment, Journal of Vocational Behavior 14 (1979) 224–247.

[37] C. Park, T. Lee, Information direction, website reputation and eWOM effect: amoderating role of product type, Journal of Business Research 62 (1) (2009)61–67.

[38] A.H. Segars, V. Grover, Re-examining perceived ease of use and usefulness: a con-firmatory factor analysis, MIS Quarterly 17 (4) (1993) 517–525.

[39] T. Sun, S. Youn, G.H. Wu, M. Kuntaraporn, Online word-of-mouth (or mouse): anexploration of its antecedents and consequence, Journal of Computer-MediatedCommunication 11 (4) (2006) Article 11.

Please cite this article as: C.M.K. Cheung, M.K.O. Lee, What drives consumplatforms, Decis. Support Syst. (2012), doi:10.1016/j.dss.2012.01.015

[40] H. Tajfel, J.C. Turner, The social identity of intergroup behavior, in: S. Worchel,W.G. Austin (Eds.), Psychology of Intergroup Relations, Nelson Hall, Chicago, IL,1986.

[41] Y. Tong, X. Wang, H.H. Teo, Understanding the intention of information contribu-tion to online feedback systems from social exchange and motivation crowdingperspectives, Proceedings of Hawaii International Conference on System Sci-ences, Hilton Waikoloa Village, Big Island, 2007.

[42] M.M. Wasko, S. Faraj, ‘It is what one does’: why people participate and helpothers in electronic communities of practice, The Journal of Strategic InformationSystems 9 (2000) 155–173.

[43] M.M. Wasko, S. Faraj, Why should I share? Examining social capital and knowl-edge contribution in electronic networks of practice? MIS Quarterly 29 (1)(2005) 35–57.

[44] B.D. Weinberg, L. Davis, Exploring theWOW in online auction feedback, Journal ofBusiness Research 58 (11) (2005) 1609–1621.

Christy M.K. Cheung is Associate Professor at Hong KongBaptist University. She received her PhD from City Univer-sity of Hong Kong. Her research interests include virtualcommunity, knowledge management, social computingtechnology, and IT adoption and usage. Her research arti-cles have been published in MIS Quarterly, Decision Sup-port Systems, Information & Management, Journal of theAmerican Society for Information Science and Technology,and Information Systems Frontiers. Christy received theBest Paper Award at the 2003 International Conferenceon Information Systems and was the PhD fellow of 2004ICIS Doctoral Consortium.

thA

Matthew K.O. Lee is Chair Professor of Information Systems& E-Commerce at the College of Business, City University ofHong Kong (CityU). Concurrently, he directs the University'sCommunication and Public Relations Office. Professor Lee'spublications in the information systems and electronic com-merce areas include a book as well as over one hundredrefereed articles in international journals, conference pro-ceedings, and research textbooks. He is the Principal Investi-gator of a number of CERG grants and has published inleading journals in his field (such as MIS Quarterly, Journalof MIS, Communications of the ACM, International Journalof Electronic Commerce, Decision Support Systems, Infor-mation & Management, and the Journal of International

Business Studies). His work has received numerous citations in the SCCI/SCI databaseand Google Scholar. Professor Lee has served as Associate Editor and Area Editor of theJournal of Electronic Commerce and Applications (Elsivier Science) and the InternationalJournal of Information Policy and Law (Inderscience) and served on the editorial board ofe Information Systems Journal (Blackwell Scientific). He has also served as a specialssociate Editor for MIS Quarterly.

ers to spread electronic word of mouth in online consumer-opinion