The Effects of Social Media

otru

dile M

omer be es ond. Tia hns,

tices. We show that brand trust has a full mediating role in converting value creation practices into brand

about,er, Schu, Zhaand reGuinn,ities in

many companies to participate in such spaces (Kaplan & Haenlein,2010).

Taking advantage of the capabilities of both brand communityand social media, several companies are using social networking

ated in brand communities. Building on these studies plus thebrand trust and loyalty literatures, we extend the concept of brandcommunity to social media and nd evidence of the three commu-nity markers in brand communities established on social mediawebsites. Then we show how these communities create valueand how they affect brand loyalty through brand trust.

Due to the rapid growth of social media and the motivation ofbrand owners to participate in social media, our purpose is toexplore whether brand communities based on social media (i.e.,a special type of online brand communities) have positive effects

Corresponding author. Tel.: +1 514 848 2424x2942; fax: +1 514 848 4576.E-mail addresses: [email protected] (M. Laroche), m_rhabibi@jmsb.

concordia.ca (M.R. Habibi), [email protected] (M.-O. Richard), [email protected] (R. Sankaranarayanan).

Computers in Human Behavior 28 (2012) 17551767

Contents lists available at

Computers in Hu

evi1 Tel.: +1 514 738 3520.learning customer perceptions of new product offerings and com-petitive actions; maximizing opportunities to attract and collabo-rate closely with highly loyal consumers of the brand (Franke &Shah, 2003; McAlexander et al., 2002); inuencing members eval-uations and actions (Muniz & Schau, 2005); rapidly disseminatinginformation (Brown, Kozinets, & Sherry, 2003; Jin, Cheung, Lee, &Chen, 2009); and most importantly gaining a holy grail of loyalcustomers (McAlexander et al., 2002). On the other hand, the dra-matic popularity and inherent advantages of the vast reach, lowcost, high communication efciency of social media are tempting

communities. Since these brand communities are becoming moreimportant and prevalent, it is necessary for both marketers andresearchers to have more insights about them.

Muniz and OGuinn (2001) found evidence of brand communi-ties in both face-to-face and computer-mediated environments.Their ethnographic study identied the presence of the threetraditional markers of community (i.e., shared consciousness,shared rituals and traditions, and obligations to society) withinbrand communities. In their meta-analysis, Schau et al. (2009)identied four groups of practices through which value is co-cre-1. Introduction

Marketers are very eager to learnbrand communities (e.g., McAlexandSchau, Muniz, & Arnould, 2009; Zhowhich include a series of connectionsple who admire a brand (Muniz & Ohind such interest in brand commun0747-5632/$ - see front matter 2012 Elsevier Ltd. Ahttp://dx.doi.org/10.1016/j.chb.2012.04.016loyalty. Implications for practice and future research opportunities are discussed. 2012 Elsevier Ltd. All rights reserved.

organize, and facilitateouten, & Koening, 2002;ng, Su, & Zhou, 2011),lationships among peo-2001). The reasons be-clude the advantages of

sites to support the creation and development of brand communi-ties (Kaplan & Haenlein, 2010; Muniz & OGuinn, 2001). Brandssuch as Jeep with famous ofine brand communities (Schauet al., 2009) try to connect with their customers and enhance theirbrand communities on such social networking sites as MySpaceand Facebook. However, despite research on ofine brand commu-nities, few studies deal with online brand communities, and espe-cially the new phenomenon of social media based brandCommunity markersBrand trust

practices (i.e., social networking, community engagement, impressions management, and brand use).Such communities could enhance brand loyalty through brand use and impression management prac-The effects of social media based brand cmarkers, value creation practices, brand

Michel Laroche , Mohammad Reza Habibi, Marie-ODepartment of Marketing, John Molson School of Business, Concordia University, 1455 d

a r t i c l e i n f o

Article history:Available online 26 May 2012

Keywords:Brand communitySocial mediaBrand loyaltyValue creation practices

a b s t r a c t

Social media based brand carticle, we explore whethcommunities) have positivthe communities as well arespondents was conducteestablished on social medshared rituals and traditio

journal homepage: www.elsll rights reserved.mmunities on brand communityst and brand loyalty

e Richard 1, Ramesh Sankaranarayananaisonneuve West, Montral, Qubec, Canada H3G 1M8

munities are communities initiated on the platform of social media. In thisrand communities based on social media (a special type of online brandffects on the main community elements and value creation practices inbrand trust and brand loyalty. A survey based empirical study with 441he results of structural equation modeling show that brand communitiesave positive effects on community markers (i.e., shared consciousness,and obligations to society), which have positive effects on value creation

SciVerse ScienceDirect

man Behavior

er .com/locate /comphumbeh

and opinions. This denition is also supported by Rheingold(1993) who postulates social media to be the social aggregations

mathat emerge from the Net when enough people carry on those pub-lic discussions long enough, with sufcient human feelings, to formwebs of personal relationships in Cyberspace.

To grasp the concept of social media two related concepts arefundamental: Web 2.0 and User Generated Content (UGC). Web2.0 was mainly used in 2004 to describe a new method in whichapplication programmers and end-users could utilize the WorldWide Web. It means that Web 2.0 is a platform in which softwareand content are not produced and published by individual compa-nies and people, but are produced and developed by differentparticipants in a continuous and collaborative manner. UGC repre-sents all the ways by which users create content and use socialmedia on the technological ground of Web 2.0. UGC, whichachieved broad popularity in 2005, is like an umbrella term thatcontains various forms of media content that are publicly availableand created by end-users from text to video and audio materials(Kaplan & Haenlein, 2010). Based on these denitions, social mediacontains a broad array of Internet-based applications but it ismostly coined with applications such as YouTube, Wikipedia,on the main community elements and on community value crea-tion practices, as well as on brand loyalty. Toward this goal, werst develop a nomological network of relationships betweenbrand community markers, value creation practices, brand trust,and brand loyalty. Then, we test the network and hypothesesquantitatively by estimating a structural equationmodel using sur-vey data from a sample of social media website users who aremembers of different brand communities on the social media web-sites. We develop and validate new scales to measure some com-munity related constructs such as shared consciousness, ritualsand traditions, and obligations to society, which may be useful inconducting future survey-based brand community studies.

The remainder of this article is structured as follows. We beginwith a brief review of the theoretical background of important con-cepts such as brand community and social media. Then we developour hypotheses and model, followed by testing the hypotheses andreporting the results. We conclude with a discussion of managerialsignicance, theoretical and practical implications, and limitationsand avenues for future research.

2. Theoretical background

2.1. Social media

People spend more than one third of their waking day consum-ing social media (Lang, 2010). Facebook alone, the hallmark of so-cial media, has over 800 million active users. The unique aspects ofsocial media and its immense popularity have revolutionized mar-keting practices such as advertising and promotion (Hanna, Rohm,& Crittenden, 2011). Social media has also inuenced consumerbehavior from information acquisition to post-purchase behaviorsuch as dissatisfaction statements or behaviors (Mangold & Faulds,2009) and patterns of Internet usage (Ross et al., 2009). These dif-ferences and importance are such that researchers are urged totreat social media as a distinct research area (Hu & Kettinger,2008).

Social media is a group of internet based applications thatbuilds on the ideological and technological foundations of Web2.0, and it allows the creation and exchange of user-generated con-tent (Kaplan & Haenlein, 2010, p.61). It is dened as the onlinetechnologies and practices which people use to share knowledge

1756 M. Laroche et al. / Computers in HuFacebook, and Second Life.Due to the advantages of social media in connecting businesses

to end-consumers directly, in a timely manner and at a low cost(Kaplan & Haenlein, 2010), in inuencing customer perceptionsand behavior (Williams & Cothrell, 2000), and in bringing togetherdifferent like-minded people (Hagel & Armstrong, 1997; Wellman& Gulia, 1999), it has become the center of attention in differentindustries. The much higher level of efciency of social media com-pared to other traditional communication channels promptedindustry leaders to state that companies must participate in Face-book, Twitter, MySpace, and others, in order to succeed in onlineenvironments (Kaplan & Haenlein, 2010). Thus, more industriestry to benet from social media as they can be used to developstrategy, accept their roles in managing others strategy or followothers directions (Williams & Williams, 2008).

In joining social media, people fulll their need for belonging-ness and their need for cognition with those who have sharednorms, values and interests (Gangadharbhatla, 2008; Tardini &Cantoni, 2005). Feeling of being socially connected is stated as acentral element in ones psychological sense of community (Sarason,1974). Although it is asserted that these feelings of belongingnesswould be weaker due to the absence of physical co-presence andlack of proximity in such environments (Constant, Sproull, &Kiesler, 1996; Granovetter, 1973), it has been shown that weak tiesare capable of bridging people (Wellman, 1997) and encouragingmembers engagement in communities to reach deep and intimatelevels (Tardini & Cantoni, 2005). This is supported by Rheingold(1991, p. 415) who stated: People are going to do what people al-ways do with a new communication technology: use it in waysnever intended or foreseen by its inventors, so turn old social codesinside out and make new kinds of communities possible. Basi-cally, taking advantage of the strong capabilities of social media,people contribute, create and join communities to fulll needs ofbelongingness, being socially connected and recognized or simplyenjoying interactions with other like-minded members. So, socialmedia and community are concepts that should be exploredtogether.

2.2. Brand community

Relationship marketing, which focuses on creating andmaintaining long term customer relationships instead of emphasiz-ing individual transactions, was introduced as a competitive advan-tage and strategic resource for the rm (Webster, 1992). Sincemaintaining one-on-one relationshipswith customers is not alwaysefcient and easy to manage, brand communities were introducedas a solution for serving customers. Brand communities executeimportant tasks on behalf of the brand. For instance, brand commu-nities support sharing information, perpetuating the history andculture of the brand, and providing assistance to consumers. Brandcommunities also provide social structure to customermarketerrelationships and greatly inuence customer loyalty (Muniz &OGuinn, 2001).

A brand community is a specialized, non-geographically boundcommunity, based on a structured set of social relations amongadmirers of a brand (Muniz & OGuinn, 2001, p. 412) and it repre-sents a form of association embedded in the consumption contextpositioned around one product. These communities are composedof its member entities and their relationships and are identied bytheir commonalities which help people share essential resources,i.e. cognitive, emotional or material in nature. However, the mostimportant entity shared among members of such community hap-pens to be the creation and negotiation of meaning (McAlexanderet al., 2002, p. 38).Many scholars studied this topic conceptually andempirically to understand the dimensions and factors shaping suchcommunities (e.g., Granitz & Ward, 1996; Holt, 1995; Muniz &

n Behavior 28 (2012) 17551767OGuinn, 1996; Muniz & OGuinn, 2001).Similar to social media, consumers have their own motivations

for joining brand communities. Brands fulll important psychological

sites to support the creation of brand communities (Kaplan &

little high talk and a lot of idle talks (Lenhart & Madden, 2007;

umaand social needs by expressing who a person is and what group theperson aligns oneself with (Elliott & Wattanasuwan, 1998). It isbelieved that consumers join brands communities to identifythemselves with brands so that their social needs of being identi-ed as persons with appropriate self-identity are met. In theirown ways, consumers search for the symbols or signs in the com-munities which help them decipher who they want to be and howthey really want to be identied by others. Pierces semiotic ele-ments constitute an iconic interrelationship where the sign resem-bles the objects (Grayson & Martinec, 2004). Schembri, Merrilees,and Kristiansen (2010) show that identication of this type of in-ter-relationships is aspirational since consumers want self-associ-ation with signs which are desirable to them. In consuming brandssuch as Ferrari and BMW, the interpretant assumes the symbolicmeaning as a part of self. Furthermore, brand communities havea potential advantage of gathering customers together and backinto conversations, which enable them to obtain information aboutthe brand from various sources (Szmigin & Reppel, 2001). Compa-nies also have incentives to enhance and support brand communi-ties. Today, more rms are realizing the advantages of online brandcommunities, which include the opportunity for effective commu-nications with their customers and obtaining valuable information.These communities not just provide an additional communicationchannel, but a possibility of establishing linkages with devotedusers (Anderson, 2005). Customers could be a valuable source ofinformation for companies. Von Hippel (2005) posits that alreadyexisting customers can be a rich source of innovative ideas andthereby could lead to a chain of actions that might produce theright product or product modications. Companies like Dell andCisco Systems transformed suppliers and customers into membersof their corporate communities, thereby enabling exchanges ofinformation and knowledge. Furthermore, Schau et al. (2009) iden-tied practices in brand communities that create value for bothcompanies and consumers. McAlexander et al. (2002) show thatthrough participation in brand-fests, companies are able to manageand increase the feelings of integration into the brand community(Jeep) and positive feelings about the brand and product category.It has also been shown that participants derive social and hedonicvalues which they cherish as an additional benet for their efforts.According to gift giving theory (Sherry, 1983), customers perceivethese social and hedonic values as a gift, without pushing for directreciprocity, thereby a sense of indebtedness and goodwill on theirpart would be enhanced that eventually increases their loyalty.

The initial concepts of ofine brand communities had a geo-graphical constraint, in that they needed customers to be physi-cally present, and were the main backbone of research(McAlexander et al., 2002; Muniz & OGuinn, 2001). Today, the con-cept of brand community is tied with media. Brands transcendgeography because media transcend geography (Muniz & OGuinn,2001), and technology made the constraint of geography almostirrelevant. The use of mobile phones, internet, and television madepeople closer than ever before. The role of social media thereforeholds the key for brands to get existing and potential customers to-gether in order to create an environment which nourishes the bondbetween customer and brand and to nd new ways to foster thisrelationship, all without the constraint of geography (Kang, Lee,Lee, & Choi, 2007). Therefore, with the emergence of new technol-ogies, internet and social media, it is imperative for businesses andcustomers to nd ways to exploit the benets of both technologyand brand community. This results in establishing online brandcommunities.

2.3. Social media based brand communities

M. Laroche et al. / Computers in HThe combination of both brand community and social medialeads to a concept that we call social media based brand communityRheingold, 1991). The communities that are formed for the pur-pose of sharing information rather than for mere commercial rea-sons exert the greatest inuence on members opinions andpurchase intentions (Algesheimer, Dholakia, & Herrmann, 2005).

Unlike traditional media where individuals consume contentpassively, content is created by community members through ac-tive participation in social media based brand communities. Thisactive creation of content shapes the character of the communityand also determines the inuences of users and participants oneach other (Bagozzi & Dholakia, 2002; Werry, 1999). Moreover,since digital environments facilitate the archiving of past contentinexpensively, these communities come to represent an aggrega-tion of collective expertise on individual topics, difcult to matchelsewhere, and create a capital of knowledge, increasing its valuefor all members. Next, we explore the effects of social media basedbrand communities on community elements, value creationpractices, brand trust and loyalty.

3. Development of the model and the hypotheses

3.1. Effects of social media based brand community on communitycommunalities

Community has been a topic of debate among social scientistsand philosophers of the nineteenth and early twentieth centuryand contemporary contributors (Dewey, 1927; Durkheim [1893]1933; Lasch, 1991). After a review of the sociology literature, Mu-niz and OGuinn (2001) identify three core components or markersof a community; shared consciousness of kind, shared rituals andtraditions, and moral responsibility or obligations to society.

3.1.1. Shared consciousness of kindFor Guseld (1978), one important element in a community is

consciousness of kind which is the shared intrinsic connectionfelt among community members. It is a way of thinking that isgreater than shared attitudes or perceptions and reects a collectivesense of difference from other members not in the same commu-nity. Weber (1978) describes it as a shared knowing of belonging.Muniz and OGuinn (2001) found thatmembers of a community feltHaenlein, 2010; Muniz & OGuinn, 2001). Brands such as Jeep withfamous ofine brand communities (Schau et al., 2009) try to con-nect with their customers and enhance their brand communitieson social networking sites such as MySpace and Facebook. Thesesites provide the user with a sense of freedom and allow them toconverse in various languages, topics and issues, which foster anenvironment that allows for the free ow of information. Peoplejoin them and other such sites and use words, videos, pictures,and avatars to stay in touch with their friends, make new friends,make plans with them, have fun and irt with them, exchangepleasantries and argue, engage in intellectual discourse, conductcommerce, exchange knowledge, share emotional support, brain-storm, gossip, feud, fall in love, nd friends, play games, create awhich is a subset of the broader concept of virtual communitiesor online brand communities; but the main differentiator is theirplatforms. The core platform of social media is Web 2.00 plus UserGenerated Content.

Originally brand communities were established on the Web 1.0platform, on companies portals or initiated by customers (Jang,Olfman, Ko, Koh, & Kim, 2008). However, with the popularity of so-cial media websites, several companies are using social networking

n Behavior 28 (2012) 17551767 1757a strong connection with one another and they termed it as theysort of know each other even if they have never met. This is thecentral facet of a brand community. They found that through

ined community. Especially the role of computer mediated

face-to-face contexts. Community members reinforce culture, his-

matory, and rituals of the community through processes of celebra-tion of brand history, storytelling, advertising, and sharedexperiences. Also, most virtual communities create and use sharedconventions and language (e.g., jargon, emoticons, or acronyms),maintain social roles, establish boundaries, enact rituals, showcommitment to communal goals, and follow norms of interaction.

3.1.3. Moral responsibilities/obligations to societyThis community marker is a felt sense of duty or obligation to

the community as a whole as well as to its members (Muniz &OGuinn, 2001). Moral responsibility has two main functions incommunities which make it very important. First it supports com-munity survival through reasoning and recognizing what is rightand what is wrong in the community, which results in integratingand retaining members. Second, moral responsibility makes mem-bers seek out help from each other in their consumption behaviorby disseminating information. Muniz and OGuinn (2001) andMcAlexander et al. (2002) found that moral responsibility mani-fested itself in their communities. Furthermore, reasoning and dis-communications such as the World Wide Web is very important.They also found several websites that echoed the feeling of sharedconsciousness of kind such as made by Saaber for anotherSaaber. . . to enjoy, and Saab Spirit, or The cult of Macintosh.This demonstrates that members are able to identify the signicantsocial category of Saab or Macintosh.

3.1.2. Shared rituals and traditionsThese are one of the factors which unite the members of a com-

munity and represent vital social processes that bring and keepthem together. This helps the meaning of the community be trans-mitted within and beyond the community (Muniz & OGuinn,2001). Rituals and traditions are a symbolic form of communica-tion that, owing to the satisfaction that members of the communityexperience through its repetition, is acted out over time in a sys-tematic fashion. Through their special meaning and their repetitivenature, rituals contribute signicantly to the establishment andpreservation of a communitys collective sense of self, which playsa role in building community identity. Rituals stabilize this identityby clarifying expected roles, delineating boundaries within andwithout community, and dening rules so that members knowthat this is the way our community is (Wolin & Bennett, 1984).The inculcation of history keeps communities vital and their cul-ture alive. Celebrating and admiring the history of a brand commu-nity helps distinguish between a true believer and theopportunistic one (Muniz & OGuinn, 2001).

Muniz and OGuinn (2001) and McAlexander et al. (2002) foundshared rituals and traditions between brand community membersand marketers both in computer mediated environments and inprocesses of legitimacy and oppositional brand loyalty members ofa brand community perpetuate their consciousness of kind. Theformer refers to the process of differentiating between the truemembers of a community and those who are not, and the latter re-fers to the process through which members of a brand communityidentify what the brand is not and who the brand communitymembers are not. These notions are also supported by otherresearchers (Englis & Solomon, 1997; Wilk, 1996).

McAlexander et al. (2002) and Muniz and OGuinn (2001) foundevidence of shared consciousness in brand communities. Moreimportantly, these feelings of oneness exceed geographical bound-aries which show that members felt belongingness to a large imag-

1758 M. Laroche et al. / Computers in Huseminating information are highly signicant in computermediated environments in which capabilities of media supportsharing information.Thus, ethnographic evidence supports that community commu-nalities are enhanced in brand communities. We are looking forsupport that such communalities are positively affected in brandcommunities established on social media, and believe that themain role of social media is to connect people and facilitate sharinginformation and meaning. As Rheingold (1991) stated people usethe new technology to do what they always did. Furthermore,our approach is quantitative and different from previous ones. Sowe propose the following hypotheses:

H1. Social media based brand community positively inuences:(a) consciousness of kind among community members, (b) sharedrituals and traditions, and (c) sense of moral responsibility.

3.2. Value creation practices

A new stream in the literature regards customers not as mererecipients of the products and values of the rm but as co-creatorsof value, competitive strategy and the rms innovation processes(Franke & Piller, 2004; Schau et al., 2009). With a meta-analysis,Schau et al. (2009) identied four categories of practices throughwhich customers co-create value in brand communities: social net-working, impression management, community engagement, andbrand use. They assert that people in brand communities are in-volved in such value creation practices that work together bothto enhance the value people realize and to promote the collectivehealth and welfare of social bodies centered on the brands. We be-lieve that brand communities, manifested by their three markers,positively inuence these value creation practices. In fact, we arguethat the three community communalities positively affect valuecreation practices especially in the context of social media whichenhance communication, information dissemination and the capa-bility of members to be in touch with each other. We elaboratemore on these issues.

3.2.1. Social networking practicesThey focus on creating, enhancing, and sustaining ties among

brand communitymembers (Schau et al., 2009).Welcoming, empa-thizing and governing are different social networking practices,which enhance similarities among members and homogeneity ofbrand communities (Schau et al., 2009). We believe social network-ing practices to be fostered by the community markers (i.e., sharedconsciousness, rituals and traditions, and obligations to society).Schau et al. (2009) bring exactly the same example (eld note) forsupporting social networking practices that Muniz and OGuinn(2001) brought for supporting shared consciousness. This impliesthat these concepts are highly related; moreover, in our context,social media could foster such practices as welcoming, empathizingand governing. However, there is no empirical support for it.

3.2.2. Community engagement practicesThese reinforce members escalating engagement with the

brand community (Schau et al., 2009). It includes practices suchas milestoning and documenting important events in the brandcommunity, and evolving cohesive personal brand narratives. Inour view, the concept of engagement goes beyond community par-ticipation; it is the process of working collaboratively with relevantpartners who share common goals and interests. Although someresearchers believe that the Internet, the World Wide Web and ingeneral the new technologies make people increasingly detachedfrom meaningful social relationships and less likely to engage thecommunity as they spend more time online (Davis, 2001; Gacken-bach, 1998; Turkle, 1996), others reject this idea and say that being

n Behavior 28 (2012) 17551767wired which they refer to being connected online has the potentialto foster and build social associations and encourage communitybuilding (Dertouros, 1997). In particular the individuals who use

found to be more, rather than less, socially engaged (Shah, Holbert,& Kwak, 1999). Thus we believe that communities in the context of

ships. So if people realize the utilitarian and hedonic values of their

umasocial media have the capability to foster engagement.

3.2.3. Impression management practicesThey are activities that have an external, outward focus on cre-

ating favorable impressions of the brand, brand enthusiasts andbrand community in the social university beyond the brand com-munity (Schau et al., 2009, p. 34). It includes practices such asevangelizing and justifying through which customers preach thebrand, share good news about it and bring some arguments toencourage others to use the brand. Online communities fosterimpressionable facts about the brand through word-of-mouthcommunications and by sharing personal experiences. Consumersengage in these activities for altruistic nature or to attain higherstatus (Dichter, 1966; Gatignon & Robertson, 1986), but Kozinets,de Valck, Wojinicki, and Wilner (2010) found that these intentionsare more complex. Whatever their motivation, members of a com-munity are involved in managing impressions activities. We be-lieve that these activities are not only observable in social mediacommunities, but are enhanced by the capabilities of social media.

3.2.4. Brand use practicesThis relates to the members tendency to help other members

with newer, improved and enhanced ways to use the focal brand.These basically include the information given by one member toanother with regards to customizing the product for better applica-bility to their needs. They also relate to feelings of one member to-wards helping or assisting other members who are relatively newto the community. Brand use practices include grooming, custom-izing, and commoditizing. All these activities results in enhanceduses of the focal brand (Schau et al., 2009). Social media could fos-ter these activities through keeping in touch brand devotees andother members and facilitating their information and resourcesharing. In addition to these explanations about value creationpractices, we believe that brand community markers directly affectvalue creation practices within the brand community. Moreover,regarding the enabling role of social media in fostering and facili-tating communication, information access and networking, theembeddedness of such brand communities in social media con-texts results in the enhancement of value creation practices.

The stronger the feelings of shared consciousness, shared ritualsand traditionsandobligations to society, themoremembers of a com-munity and the company feel united, close to each other and moti-vated which enhance collaborative value creation practices. Forexample, ifmembers consider themselves highly obligated to society,they are more likely to preach and evangelize the brand, help eachother better use the branded products, or try to enhance and sustaintheir ties, all of which are value creation practices. It was also shownthat shared consciousness enhances interpersonal ties in a group andincreases the willingness to share information and resources withother members, to provide support and to commit to goals identiedby the group (Walther, 1996;Wellman, 1999). Thus, we hypothesizethat community markers positively affect value creation practices inthe context of social media based brand communities:

H2. Social networking practices are positively inuenced by the (a)shared consciousness of kind, (b) shared rituals and traditions, and(c) sense of moral obligations of members of the community.

H3. Community engagement practices are positively inuenced byinternet communities to explore interests and gather data are

M. Laroche et al. / Computers in Hthe (a) shared consciousness of kind, (b) shared rituals and tradi-tions, and (c) sense of moral obligations of members of thecommunity.brand their trust would increase. In brand communities andthrough value creation practices, people develop close relation-ships and draw values from their long term interactions, whichmakes them trust and love the brand (Carroll & Ahuvia, 2006; Zhouet al., 2011). Based on these arguments we hypothesize:H4. Impression management practices are positively inuenced bythe (a) shared consciousness of kind, (b) shared rituals and tradi-tions, and (c) sense of moral obligations of members of thecommunity.

H5. Brand use practices are positively inuenced by the (a) sharedconsciousness of kind, (b) shared rituals and traditions, and (c)sense of moral obligations of members of the community.

These hypotheses are depicted in Fig. 1, which provides themodel to be tested.

3.3. Effects of value creation practices on brand trust and brand loyalty

Brand community researchers agree that one of the main func-tions, if not the main one, of a brand community is to make cus-tomers loyal to the brand (McAlexander & Schouten, 1998;McAlexander et al., 2002; Muniz & OGuinn, 2001; Schau et al.,2009; Zhou et al., 2011). However, despite qualitative evidence itis still unclear how the process of being more loyal happens inbrand communities. Like researchers who refer to the power of on-line communities as a social structure, in building and enhancingtrust and loyalty (Ba, 2001; Walden, 2000), we argue that socialmedia based brand communities enhance brand loyalty throughthe enhancement of brand trust. In fact, value creation practices af-fect brand trust through three mechanisms, enhanced by capabili-ties of social media, and brand trust increases brand loyalty. Thus,brand trust would have a mediating role.

Brand trust is the willingness of the average consumer to relyon the ability of the brand to perform its stated function (Chaudh-uri & Holbrook, 2001, p. 82). This notion is more prominent in sit-uations of uncertainty, information asymmetry and fear ofopportunism; thus, the role of trust is to decrease uncertaintyand information asymmetry and make customers feel comfortablewith their brand (Chiu, Huang, & Yen, 2010; Doney & Cannon,1997; Gefen, Karahanna, & Straub, 2003; Moorman, Zaltman, &Deshpande, 1992; Pavlou, Liang, & Xue, 2007). One of the mecha-nisms for value creation practices to enhance trust is based oninformation dissemination. In all practices, e.g., evangelizing, cus-tomizing, welcoming, justifying and documenting (Schau et al.,2009), information passes among members from how to betteruse the branded products to personal information and storiesabout the brand. Such information sharing reduces uncertaintyand information asymmetry, and increases predictability of brandactions (Ba, 2001; Lewicki & Bunker, 1995). Furthermore, the plat-form of social media facilitates the process of information dissem-ination. In the literature, repeated interaction and long termrelationships are introduced as keys to building trust (Holmes,1991; Wang & Emurian, 2005). All value creation practices increasethe level of interactions between consumers and the product, thebrand, other customers and marketers, all of which are elementsof a brand community (McAlexander et al., 2002). The placelessand comprehensive nature of social media based brand communi-ties facilitate long term, close contacts without the loss of relation-ships that often accompanies ofine environments (Ba, 2001).These enhanced relationships results in brand trust.

Trust also involves a calculative process (Doney & Cannon,1997, p. 37) related to the value people receive from their relation-

n Behavior 28 (2012) 17551767 1759H6. Social networking practices positively inuence brand trust.

ocia

Bra

ects

maH8. Impression management practices positively inuence brandtrust.H7. Community engagement practices positively inuence brandtrust.Brand Community (on Social Media)

Shared Consciousness

Shared Rituals and Traditions

Obligation to Society

S

H1a

H1b

H1c

H2a

H5a

H3aH2b

H2c

H3b

H3c

H4b

H4c

H5b

H5c

H4a

Fig. 1. Model of the eff

1760 M. Laroche et al. / Computers in HuH9. Brand use practices positively inuence brand trust.A great deal of research supports the positive effect of trust on

loyalty in online or ofine contexts (Chaudhuri & Holbrook, 2001;Chiu et al., 2010; Harris & Goode, 2004; Kim, Chung, & Lee, 2011;Zhou et al., 2011). Brand trust is one of the most important ante-cedents of brand loyalty; however in the brand community litera-ture there is little emphasis on this relationship, so wehypothesize:

H10. Brand trust positively inuences brand loyalty.Fig. 1 shows the complete model of how brand community af-

fects brand loyalty. As depicted, brand trust has a mediating rolebetween the effects of value creation practices and brand loyalty.

4. Method

4.1. Subjects and procedure

Since we test our model in the context of social media basedbrand communities, the target population is all members of brandcommunities on social networking websites. We used the screen-ing condition that respondents had to be members of such brandcommunities. The study was introduced as an opinion surveyand participants were asked to list the name of the communitythey associate themselves with; are a member of any brand com-munity or any social networking site and to consider them whenresponding to questions. This is important because the study aimsto investigate the effects of such communities on members,regardless of a specic product, brand or social media. The ques-tionnaires were distributed through social network websites suchas Facebook, MySpace, and Twitter. Several posts were includedin these heavy trafc websites along with email distribution lists.This method of data collection, which is consistent with previousonline studies (Bagozzi & Dholakia, 2006; Steenkamp & Geyskens,2006) generated 441 valid responses (41.1% female). The age rangevaried between 18 and 55.

l Networking

Community Engagement

Impression Management

nd Use

Brand LoyaltyBrand Trust

H6

H7

H8

H9

H10

of brand community.

n Behavior 28 (2012) 175517674.2. Measures

All constructs were measured with multiple item-scales. Someitems were adopted from the related literature and modied to suitthe study, and some were developed from the literature (AppendixA). A pretest with 30 respondents was conducted to validate theconstructs.

4.2.1. Brand communityAs we measured it in a social media context, we rst asked

respondents to answer the items based on their experiences withbrand communities on social networking sites of which they aremembers. A set of six items on a 5-point Likert scale measured thisconstruct (Srinivasan, Anderson, & Ponnavolu, 2002). Based on thevalidity of this scale, we adopted it with some modications to suitour context. The Cronbachs alpha is 0.72.

4.2.2. Shared consciousness, shared rituals and traditions, andobligations to society

The scales were derived from the denitions given by Munizand OGuinn (2001) and consisted of two items each. The itemswere tested in a three factor conrmatory model and an acceptablet was obtained, with Cronbachs alphas of 0.72, 0.73 and 0.75respectively. The items were also successfully tested for discrimi-nant validity. The average variances extracted for the scales were0.78, 0.79 and 0.65 respectively, showing convergent validity.

4.2.3. Social networkingEight 5-point Likert items measured this construct. The scale

was developed for use with an online store (Hsieh, Chiu, & Chiang,

(Algesheimer et al., 2005). The scale reliability is 0.71 and the aver-

4.2.6. Impression management

It was measured as a three-item index based on the four item

It was derived from Delgado-Ballester, Manuera-Aleman, and

5. Results

by nonsignicant chi-squared tests, RMSEA value less than 0.08

minimum of 0.70 (Cronbach, 1970; Nunnally, 1978) and the aver-

umaage variance extracted (AVE) which estimates the amount of vari-ance captured by a constructs measure relative to randommeasurement error (Fornell & Larcker, 1981). An AVE greater than0.50 is considered to support internal consistency (Bagozzi & Yi,and CFI values greater than or equal to 0.90.

5.1. Internal consistency

Two measures were used to evaluate the internal consistency ofthe constructs. Reliability is identied by Cronbachs alpha with aStructural equation modeling with the AMOS graphical soft-ware program was used to test the model and hypotheses shownin Fig. 1. The goodness-of-t of the overall model was assessedwith the chi-square test, the root mean square error of approxima-tion (RMSEA), and the comparative t index (CFI) (Bentler, 1990;Marsh, Balla, & Hau, 1996). Satisfactory model ts are indicatedYague-Guillen (2003). The three items are measured on a 5-pointLikert scale characterized by a reliability of 0.84 and an averagevariance extracted of 0.57.index originally developed by Chaudhuri and Holbrook (2001)using a 5-point rating of agreement with three statements: I trustmy brand to give me everything I expect out of it, I rely on thisbrand, and My brand never disappoints me. The Cronbachs al-pha for this index is 0.71.

4.2.8. Brand loyaltyIt is derived from the denition given by Schau et al. (2009). Thescale consists of three 5-point Likert items. The average varianceextracted is 0.63 and the Cronbachs alpha is 0.71.

4.2.7. Brand trustage variance extracted is 0.64.

4.2.5. Brand useThree 5-point Likert-type statements measured the degree to

which a member of a community gets useful information aboutthe brands use. The items were constructed from the denitionof brand use practices given by Schau et al. (2009). The averagevariance extracted is 0.61 and the Cronbachs alpha is 0.70.2005). However, since the scale measures the same construct as inthe original study, it was used with a slight modication. The scalehas a reliability of 0.78. Further analyses provided evidence of con-vergent validity. The average variance extracted is 0.65.

4.2.4. Community engagementFour 5-point Likert items measured this construct. The items

measured the degree to which a person is involved with a commu-nity of brand users due to intrinsic benets of the activity

M. Laroche et al. / Computers in H1988). All AVEs are greater than 0.50. Table 1 also shows descrip-tive statistics for the constructs and their reliability measureswhich are all above 0.70.5.2. Unidimensionality

The evaluation of unidimensionality of the proposed scales wasassessed by performing exploratory factor analysis (EFA). Factorswere extracted based on eigenvalues greater than 1. In additionthey were required to have a signicant factor loading greater than0.4 and a high total variance explained component. Based on theresults only one factor was extracted for shared consciousness, rit-uals and traditions, obligations to society, impression manage-ment, community engagement, brand use, brand loyalty, brandtrust, social networking and online brand communities constructs.

A total of 10 constructs were formed based on their eigenvalues.Bartletts test of Sphericity and Kaiser-Meyer-Olkin (KMO) mea-sure of sampling adequacy are provided in Table 2. Bartletts Testof Sphericity determines whether there is a high enough degreeof correlation among the variables. Table 2 shows that the test issignicant (p < .001). The KMO stat as indicated Table 2 is 0.85which is greater than the minimum of 0.5 indicating a good degreeof correlation. Hence the test reveals a meaningful factor analysisand principal component analysis can be carried out. The total var-iance explained by the 10 factors is 66.8%.

In the EFA, we found that the items The community allows di-rect user input or posting to site (ob1), The brand communitykeeps in touch with me with notications (sn1), The communityprovides with me product information (sn4), I share my opinionson the community (sn8) and I benet from following the com-munitys rules (ce1) did not have high enough loadings on theirrespective factors. Hence they were eliminated from furtheranalysis.

5.3. CFA results

Next we conducted conrmatory factor analysis (CFA). Poor fac-tor loadings were found for two items of brand community mea-sures (ob2 with loading .05, and ob3 with loading 0.12, both areless than 0.4) so we dropped them out and run the model again.We found good t indices for the CFA model including all latentvariables assuming they are correlated (further analysis shows thisis a true assumption since correlations among all variables are sig-nicantly greater than zero; Table 3); v2 = 478.0, p = .000,RMSEA = 0.032, CFI = 0.97, GFI = 0.93, RMR = 0.024. Except thechi-square which is sensitive to sample size and not a good indica-tor, all other statistics show a good t for the CFA model (Bagozzi &Yi, 1988). Table 3 shows factor loadings and R2 for each item (seeAppendix A for the items).

5.4. Discriminant validity

Discriminant validity is achieved when the correlations be-tween the constructs differ signicantly from 1 or when chi-squaredifference tests indicate that two constructs are not perfectly cor-related. As a test of discriminant validity, the correlations amongthe latent variables were checked to determine if they are signi-cantly different from 1. Table 4 shows the result for the discrimi-nant validity, conrming the validity of all constructs.

5.5. Convergent validity

Referring to the CFA model, since all the factor loadings areabove 0.5, convergent validity is supported (Steenkamp & Geys-kens, 2006).

5.6. Structural model estimation

n Behavior 28 (2012) 17551767 1761With respect to the t statistics for the full model(v2(347) = 572.8, v2/df = 1.65, p = 0.000), the chi-square is

7.91 2.24 0.715.91 1.75 0.71

maTable 1Means, standard deviations, reliability statistics for construct measures.

Constructs No. of items

Social media brand community 6Social networking 8Community engagement 4Impression management 3

1762 M. Laroche et al. / Computers in Husignicant (p < 0.05), which is usually the case for large samplesizes, and the v2/df is less than 2. All other statistics are withinacceptable ranges including RMSEA = 0.038, CFI = .95, GFI = 0.92,IFI = 0.95, RMR = 0.028 (Bagozzi & Yi, 1988).

We found strong support for the effects of social media basedbrand community on the three markers of community, i.e., sharedconsciousness, shared rituals and traditions, and obligations tosociety. The b values are respectively: 1.27, 1.30, and 1.05. All ofthese relationships are signicant at the 0.001 level, providing

Brand use practices 3Shared consciousness 2Obligations to society 2Rituals and traditions 2Brand loyalty 3Brand trust 3

Table 2KMO and Bartletts test.

Kaiser-Meyer-Olkinmeasure of sampling adequacy

.849

Bartletts test of sphericity Approx. chi-square 2580.26df 630Sig. .000

Table 3Factor loadings and R-square of the items in CFA.

Construct Item Factor loading R-square

Brand community Ob4 0.491 0.241Ob5 0.657 0.431Ob6 0.724 0.525

Social networking Sn2 0.643 0.413Sn3 0.688 0.473Sn5 0.629 0.396Sn6 0.572 0.327Sn7 0.613 0375

Community engagement Ce2 0.531 0.282Ce3 0.653 0.426Ce4 0.709 0.502

Shared consciousness Sc1 0.750 0.562Sc2 0.751 0.563

Impression management Im1 0.640 0.410Im2 0.696 0.485Im3 0.676 0.457

Brand trust Bt1 0.631 0.398Bt2 0.569 0.323Bt3 0.560 0.314

Loyalty L1 0.797 0.636L2 0.854 0.730L3 0.803 0.645

Obligations to society Obsoc1 0.744 0.553Obsoc2 0.757 0.573

Rituals and traditions Tt1 0.806 0.649Tt2 0.715 0.511

Brand use Bu1 0.602 0.363Bu2 0.637 0.405Bu3 0.739 0.545

Note: See Appendix A for the items.support for H1a, H1b and H1c. Fig. 2 summarizes these and otherresults.

Shared consciousness has a signicant effect on social network-ing practices (b = 0.29, p < .01) supporting H2a. It has a positive,signicant effect on community engagement and brand use prac-tices at p < .05 level, with b = 0.21 and b = 0.14, respectively, whichsupport H3a and H5a. However, the effect of shared consciousnesson impressions management (H4a) is not supported b = 0.07,p = 0.36.

The effects of shared rituals and traditions on value creationpractices were hypothesized in H2b, H3b, H4b and H5b. We foundsupport for all these hypotheses at p < 0.01. The effects of sharedrituals and traditions are positive and signicant on: social net-working (b = 0.33), community engagement (b = 0.37), impressionsmanagement (b = 0.33), and brand use practices (b = 0.26).

As expected, obligations to society has signicant positive inu-ences on all the four constructs of value creation practices (socialnetworking, community engagement, brand use practices, andimpressions management) in support of H2c, H3c, H4c and H5cwith signicance levels of p < 0.001, and b values of 0.40, 0.47,0.43 and 0.39 respectively.

The effects of value creation practices on brand trust werehypothesized in H6 to H9. The effects of social networking andcommunity engagement practices on brand trust (H6 and H7) arenot supported with bs (p-values) respectively: 0.06 (0.65), and0.12 (0.28). However, the effect of impression management onbrand trust (H8) is signicant at p < 0.001 with b = 0.31. The effectof brand use practices on brand trust (H9) is signicant (b = 0.29,p < .05). As predicted by H10, the effect of brand trust on brand loy-alty is signicant and positive (b = .86, p < .001).

The ndings show that brand trust fully mediates the effect ofbrand use and impressions management practices on brand loy-alty. To conrm that brand trust fully mediates these effects wetested another model with brand trust as a partial mediator. Weadded direct relationships from the four value creation practices

5.95 1.72 0.703.96 1.34 0.723.84 1.16 0.753.95 1.55 0.736.73 2.72 0.865.34 1.70 0.71Mean Standard deviation Cronbachs a

10.83 2.76 0.7215.8 4.08 0.78

n Behavior 28 (2012) 17551767to brand loyalty in the current model (Fig. 3).We tested this model, but none of the new relationships were

signicant and the model t did not improve (v2 = 570.4,df = 343, RMSEA = 0.039, CFI = 0.95, GFI = 0.92, IFI = 0.95). This im-plies that brand trust fully mediates the effects of the two valuecreation practices, i.e., brand use and impressions management,on brand loyalty.

6. Discussion and implications

This study looked at the fast growing phenomenon of brandcommunities established on social media. Drawing on the litera-ture on brand community, we proposed a model of the effects ofbrand community on shared community markers, value creationpractices and the way they convert to brand loyalty through brandtrust. Using SEM we found support for the model and most of our

B =

umaTable 4Correlation matrix.

SC OBSOC TT OB

SCOBSOC .419

TT .388 .430

OB .465 .481 .454

SN .538 .640 .589 .635

CE .488 .584 .540 .554

BU .424 .499 .430 .501

IM .377 .484 .473 .456

LL .339 .273 .424 .266

BT .393 .471 .453 .456

Notes: All coefcients are signicant at p < 0.01.SC = shared consciousness, OBSOC = obligation to society, TT: rituals and traditions, Ouse, IM = impressions management, LL = brand loyalty, BT = brand trust.

M. Laroche et al. / Computers in Hhypotheses. We found that brand communities established onsocial media enhance feelings of community among membersand contribute to creating value for both members and the com-pany. Furthermore, the model shows how brand loyalty is in-creased in brand communities. The whole model shows thatvalue creation practices enhance loyalty through brand trust as amediator.

Contrary to expectations, only two of the four value creationpractices (brand use and impressions management) contribute tobrand trust. There is one possible explanation for this nding. AsSchau et al. (2009, p. 37) argue, Practices operate like apprentice-ship, which means that their effects evolve over time as con-sumer engagement deepens and practices are integrated. Thus,it is possible that the effects of practices in social media basedbrand communities may not have evolved enough to signicantlyaffect brand loyalty. This makes sense because the concept andprevalence of social media is new and its effects are emerging astime goes on and more people engage in the activities of socialmedia. This difference might be due to the difference in the natureof practices as well. For example the effects of brand use practices,

Brand Community (on Social Media)

Shared Consciousness

Shared Rituals and Traditions

Obligation to Society

Soci

1.27 **(.163)

1.3**(.173)

1.045**(.142)

.292**(.063)

.140*(.065)

.210*(.08)

.325**(.061)

.400**(.071)

.366**(.079)

.472**(.092).325**

(.074)

.425**(.087) .264**

(.066)

.389**(.079)

Fig. 2. Estimated model. p < .001, p < .05. Note: Unstandardized coefcients are used anSN CE BU IM LL

.644

.564 .554

.540 .540 .530

.317 .299 .412 .378

.463 .468 .471 .485 .467

brand community, SN = social networking, CE = community engagement, BU = brand

n Behavior 28 (2012) 17551767 1763which mainly are about better consumption of the focal brand, aremore immediate than social networking practices that need moretime to develop.

The effect of shared consciousness on impressions management(H4a) was not signicant as well. This has two possible explana-tions. First, it is possible that merely having a shared consciousnessor feeling of a community does not lead to actively preaching andevangelizing the community and its ethos. For doing so, consumersneed to have more incentives and be more obligated to society. Asour results show, feelings of obligations to society have signicantand strong effects on impressions management. This makes moresense considering the fact that having a large number of brandcommunity members in social media weakens the ties amongusers. Second, Muniz and OGuinns (2001) legitimacy, an elementof shared consciousness, exists for some brand communities butnot for others. For example, they did not nd evidence of legiti-macy for the Macintosh community. Since we aggregated datafrom different communities, it is possible that existence and non-existence of such elements in different communities neutralizedeach others effects so that we do not have large enough effect sizes

al Networking

Community Engagement

Impression Management

Brand Use

Brand LoyaltyBrand Trust

.313**(.083)

.290*(.114)

.858**(.129)

d standard errors are in parentheses; dashed arrows represent nonsignicant paths.

Netw

and U

and

mato reach signicance. Focusing on specic brand communitiesbased on social media will address this issue, which is an avenuefor future research.

Our study contributes to existing brand community knowledge

Brand Community (on Social Media)

Shared Consciousness

Shared Rituals and Traditions

Obligation to Society

Social

Br

.119 **(.001)

.495**(.094)

.072**(.008)

2.021**(.434)

1.170*(.508)

.754*(.806)

.584**(.001)

3.584**(.587)

.896**(.227)

6.53**(.319).755**(.171)

5.83**(.978)

.528**(.129)

4.170.**(.743)

Fig. 3. The model with brand trust as partial mediator. p < .001, p < .05. Note: Unstrepresent nonsignicant paths.

1764 M. Laroche et al. / Computers in Huin several ways and has theoretical implications. First, we extendedthe concept of brand community to social media and created moreinsights on this important phenomenon. Recently, the uniquenessof social media has been acknowledged by researchers as a distinctresearch domain (e.g., Hu & Kettinger, 2008), so our study providesmore insights into this domain. Second, we developed a nomolog-ical network that shows how a community affects communitymarkers, practices, brand trust and brand loyalty; then we testedit in the context of social media. Although, previous researchersshowed qualitatively the existence of such effects in brand com-munities (McAlexander et al., 2002; Muniz & OGuinn, 2001), wemodeled these effects in online environments, and tested the mod-el quantitatively. Our ndings explicitly show how practices couldaffect brand loyalty as well as their relative weights. For example,we show that brand communities affect loyalty through brand usepractices. However, in interpreting the results we should be cau-tious because of the apprenticeship effects of such practices. Third,we created new scales to measure some of the constructs whichcould be used by other researchers.

Thisworkhas important practical implications formanagers. Thevast reach, low cost and popularity of socialmedia are prompting allbrand managers to take advantage of such characteristics, so thatthey, almost blindly, want to be on social media. Our ndings helpthem havemore insights into this process. First, it shows that brandcommunities based on social media have the same advantages andbenets as ofine brand communities. They positively inuencethe shared consciousness, shared rituals and obligations to societyof community members. Furthermore, they result in enhancing va-lue creation practices. Second, our ndings show that if managerswish to enhance brand loyalty, they have to promote brand useand impressionsmanagementpractices in their brand communities.Brand use practices refer to disseminating information or otheractivities that help members more optimally use their brandedproduct; impressions management involves practices that manageexternal impressions and images of the brand (Schau et al., 2009).Obviously these activities couldbepromotedeasilywith socialmed-

orking

Community Engagement

Impression Management

se

Brand Loyalty

Brand Trust

.134**(.081)

.200*(.084)

.669*(.310)

ardized coefcients are used and standard errors are in parentheses; dashed arrows

n Behavior 28 (2012) 17551767ia capabilities of sharing information and connecting people. An-other important observation from the model is the relativeimportance of obligations to society. The path coefcients fromobli-gations to society to value creationpractices are larger than the coef-cients of other paths. This implies that marketers could enhancevalue in communities by emphasizing obligations and commitmentto society. Themoremembers feel committed and obligated to soci-ety the more they would get involved in practices that create valuefor the community.

7. Limitations and future research

Despite these contributions, we acknowledge the limitations ofthis research and accordingly propose new avenues for research.First, we showed that brand communities based on social mediacan enhance feelings of community and value creation practices.Regarding our goals, we mainly concentrated on general notionsand concepts in the brand community and social media literatures.However, in the future researchers may consider more variablesand factors to produce deeper insights. For example, potentialmoderators and mediators such as brand type, culture, structureand type of communities as well as the differences in social mediaplatforms functionality and features (Kietzmann, Hermkens,McCarthy, & Silvestre, 2011) could be investigated.

However, it should be noted that brand communities based onsocial media might not be easy to manage. In this environment,customers are more powerful than before. They easily can interact,speak and broadcast their ideas to each other while companieswould have increasingly less ability to manage the informationavailable about them in the new space (Kaplan & Haenlein, 2010).In addition, consumers may get involved in online complaints withforming protest websites or weblogs if they are dissatised, or

up w W sde -ti h u .Ac d hco q -m i

e -ic b i em o e rtime (Schau et al., 2009 -fe h -na u r eco r

8. Conclusion

t -ties established on the p gel n tan r b -ne n o ssh d ll

consciousness

4 Rituals and

society

9 Brand loyalty

Humaestablished notions and theories in the social media contexts. Tosum up briey, from our ndings it seems appropriate to echothe optimism of brand community researchers such as McAlexan-der et al. (2002), Muniz and OGuinn (2001) and Algesheimer et al.(2005) and to suggest that social media based brand communitiesoffer brand owners the ability to enhance value, brand trust, brandloyalty and feelings of community among members. Marketersmay do well to take advantage of the opportunities that such brandcommunities present.

Acknowledgements

The authors gratefully acknowledge the nancial support of theSocial Sciences and Humanities Research Council of Canada.

Appendix A. Summary of measures

1 Brandcommunity

Ob1 The community allows directuser input or posting to siteOb2 Customers share experiencesabout products online with othercustomersOb3 The community is useful forgathering various information about theproduct or the brandOb4 The members of this communitybenet from the communityOb5 The members share a commonbond with other members of thecommunity.Ob6 The members are stronglyafliated with other members

2 Socialnetworking

Sn1 The brand community keeps intouch with me with noticationsSn2 At least some of members of mycommunity know meSn3 I received special treatment after Ibecame a memberSn4 The community provides with mess aouland loyalty. Wed rapid growthpay special attenelieve that due to the importance, uniquef social media, marketers and researchertion to this phenomenon and examine weemed bts and practicesmonstrated the effects of brand communilatform of social media on the underlyinin communities as well as on brand trusIn his article, we demmunity and compa

esearchers to follow the changes in the the results at different times.cts al stct over time anddies to enablects of value creation practices evolve ove); however; we do not know how these efow they develop. So we suggest longitudis ofentirand communitned that the effue for research is to investigate the dynames based on social media. For example, wunitOnuences and introes in such envirointeresting avenduce effective techniques to manage comnments.cornseingly, more rese

tious about their potential consequencesarch is needed to study and analyze succidees, tto establish or eney should be caard & Ostrom, 2006). Thus, as businessehance online social media based communiset ith the brand (M. Laroche et al. / Computers inproduct informationsome other storeL3 I am willing to pay more for mybrandL2 If the brand is not available at thestore I would buy the same brand fromdeemed to help community building.L1 I consider myself to be loyal to thebrandBu3 Members of my communitymonitor and foster the activitiesand unsuccessful attempts atcustomization of the productproduct or brandBu2 Members of my community sharetheir experiences about their successfulBrand use Bu1 Members of my community shareuseful tips about better uses of the8Im3 Members actively defend/refutethe actions of the companysmanagementIm2 Members actively engage indiscussions in order to justify theirreasons for their afnity towards thebranddiscussions related to company, brandor the product7 Impressionmanagementable to reach personal goalsIm1 Community encouragesable to support other membersCe4 I am motivated to participate inthe communitys activities because I amafterwards or because I like itCe3 I am motivated to participate inthe communitys activities because I amcommunitys rules.Ce2 I am motivated to participate inthe activities because I feel good6 CommunityengagementObsoc2 The community engages inintegrating and retaining members.Ce1- I benet from following theObcoc1 The members of thecommunity assist/advice othermembers of the same community in theproper use of the brand5 Obligations toTt2 I think these traditions contributetowards a specic culture of thecommunitytraditionsyour communityTt1 I recollect vital social traditions orrituals specic to the brand communitySc1 An intrinsic connection is feltamong the membersSc2 A general sense of differenceexists from members who are NOT in3 Sharedgreetings.Sn8 I share my opinions on thecommunitySn7 The community recognizesspecial occasions and sends memy needs.Sn6 The community collects myopinions about the services/productSn5- The community is concerned withn Behavior 28 (2012) 17551767 1765(continued on next page)

Franke, N., & Shah, S. K. (2003). How communities support innovative activities: An

Gangadharbhatla, H. (2008). Facebook me: Collective self-esteem, need to belongd

i oGa al

sGe D r :

iGr

Gr

mafen,Ananitz,Adva

anove136., Karahanna, E., & Stntegrated model. MIS

N. A., & Ward, J. C. (1nces in Consumer Restter, M. (1973). The s01380.in Consumer Research, 13, 534538.aub, D. W. (2003). Trust and TAM in online shoppingQuarterly, 27, 5190.tignocomn, H., & Robertson, T.munication. AdvanceS. (1986). An exchange theory model of interpersonandsocinternet self-efcacyal networking sites. Jas predictors of the I-generations attitudes towarurnal of Interactive Advertising, 8(2), 515.exploration of assistance and sharing among end-users. Research Policy, 32,157178.

Gackenbach, J. (1998). Psychology and internet: Interpersonal, intrapersonal andtranspersonal implications. San Diego: Academic Press.10 Brand trust BT1- My brand gives me everything thatI expect out of the productBT2 I rely on my brandBT3- My brand never disappoints me

Note: All responses are based on 5-point Likert scales anchored by: 1 = Totallydisagree and 5 = Totally agree.

References

Algesheimer, R., Dholakia, U., & Herrmann, A. (2005). The social inuence of brandcommunity: Evidence from European car clubs. Journal of Marketing, 69(3),1934.

Anderson, P. H. (2005). Relationship marketing and brand involvement ofprofessionals through web-enhanced brand communities: The case ofColoplast. Industrial Marketing Management, 34(3), 285297.

Ba, S. (2001). Establishing online trust through a community responsibility system.Decision Support System, 31, 323336.

Bagozzi, R., & Dholakia, U. (2002). Intentional social action in virtual communities.Journal of Interactive Marketing, 16(2), 221.

Bagozzi, R., & Dholakia, U. (2006). Open source software communities: A study ofparticipation in Linux user groups. Management Science, 52, 10991115.

Bagozzi, R., & Yi, Y. (1988). On evaluation of structural equation models. Journal ofthe Academy of Marketing Science, 16, 7494.

Bentler, P. M. (1990). Comparative t indexes in structural models. PsychologicalBulletin, 107, 238246.

Brown, S., Kozinets, R. V., & Sherry, J. F. Jr., (2003). Teaching old brands new tricks:Retro branding and the revival of brand meaning. Journal of Marketing, 67,1933.

Carroll, B. A., & Ahuvia, A. C. (2006). Some antecedents and outcomes of brand love.Marketing Letters, 17, 7989.

Chaudhuri, A., & Holbrook, M. B. (2001). The chain of effects from brand trust andbrand affect to brand performance: The role of brand loyalty. Journal ofMarketing, 65(2), 8193.

Chiu, C. M., Huang, H. Y., & Yen, C. H. (2010). Antecedents of online trust in onlineauctions. Electronic Commerce Research and Application, 9, 148159.

Constant, D., Sproull, L., & Kiesler, S. (1996). The kindness of strangers: Theusefulness of electronic weak ties for technical advice. Organization Science, 7,119135.

Cronbach, L. J. (1970). Essentials of psychological testing. New York: Harper and Row.Davis, R. A. (2001). A cognitive-behavioral model of pathological Internet use.

Computers in Human Behavior, 17(2), 187195.Delgado-Ballester, E., Manuera-Aleman, J. L., & Yague-Guillen, M. J. (2003).

Development and validation of a brand trust scale. International Journal ofMarket Research, 45(1), 3554.

Dertouros, M. L. (1997). What will be: How the new world of information will changeour lives. San Francisco: Harper Edge.

Dewey, J. (1927). The public and its problems. New York: Holt.Dichter, E. (1966). How word-of-mouth advertising works. Harvard Business Review,

16, 147166.Doney, P. M., & Cannon, J. P. (1997). An examination of the nature of trust in buyer

seller relationships. Journal of Marketing, 61, 3551.Durkheim, E. (1933). The division of labor in society, trans. George Simpson. New York:

Free Press [1893].Elliott, R., & Wattanasuwan, K. (1998). Brands as Symbolic resources for the

construction of identity. International Journal of Advertising, 17, 131144.Englis, B. G., & Solomon, M. R. (1997). I am not, therefore, I am: The role of avoidance

products in shaping consumer behavior. Advances in Consumer Research, 24,6162.

Fornell, C., & Larcker, D. F. (1981). Evaluating structural equation models withunobservable variables and measurement error. Journal of Marketing Research,28, 3950.

Franke, N., & Piller, F. (2004). Value creation by toolkits for user innovation anddesign: The case of the watch market. Journal of Product Innovation Management,21, 401415.

1766 M. Laroche et al. / Computers in Hu996). Virtual community: A socio-cognitive analysis.earch, 23, 161166.trength of weak ties. American Journal of Sociology, 78,Grayson, K., & Martinec, R. (2004). Consumer perceptions of iconicity andindexicality and their inuence on assessments of authentic market offerings.Journal of Consumer Research, 31, 296312.

Guseld, J. (1978). Community: A critical response. New York: Harper and Row.Hagel, J., & Armstrong, A. G. (1997). Net gain: Expanding markets through virtual

communities. Boston, MA: Harvard Business School Press.Hanna, R., Rohm, A., & Crittenden, V. L. (2011). Were all connected: The power of

the social media ecosystem. Business Horizons, 54, 265273.Harris, L. C., & Goode, M. M. H. (2004). The four levels of loyalty and the pivotal role

of trust: A study of online service dynamics. Journal of Retailing, 80, 139158.Holmes, J. G. (1991). Trust and the appraisal process in close relationships. In W. H.

Jones & D. Perlman (Eds.). Advances in personal relationships (Vol. 2,pp. 57104). London: Jessica Kingsley.

Holt, D. B. (1995). How consumers consume: A typology of consumption practices.Journal of Consumer Research, 22(1), 116.

Hsieh, Y., Chiu, H., & Chiang, M. (2005). Maintaining and committed onlinecustomer: A study across search-experience-credence products. Journal ofRetailing, 81, 7582.

Hu, T., & Kettinger, W. J. (2008). Why people continue to use social networkingservices: Developing a comprehensive model. In Proceedings, 29th internationalconference on information systems, Paris, France.

Jang, H., Olfman, L., Ko, I., Koh, J., & Kim, K. (2008). The inuence of online brandcommunity characteristics on community commitment and brand loyalty.International Journal of Electronic Commerce, 12(3), 5780.

Jin, X. L., Cheung, C. M. K., Lee, M. K. O., & Chen, H. P. (2009). How to keep membersusing the information in a computer-supported social network. Computers inHuman Behavior, 25, 11721181.

Kang, I., Lee, K. C., Lee, S., & Choi, J. (2007). Investigation of online community voluntarybehaviour using cognitive map. Computers in Human Behavior, 23, 111126.

Kaplan, A. M., & Haenlein, M. (2010). Users of the world, unite! The challenges andopportunities of social media. Business Horizons, 53, 5968.

Kietzmann, J. H., Hermkens, K., McCarthy, I. P., & Silvestre, B. S. (2011). Socialmedia? Get serious! Understanding the functional building blocks of socialmedia. Business Horizons, 54, 241251.

Kim, M., Chung, N., & Lee, C. (2011). The effect of perceived trust on electroniccommerce: Shopping online for tourism products and services in South Korea.Tourism Management, 32, 256265.

Kozinets, R., de Valck, K., Wojinicki, A. C., & Wilner, S. (2010). Networked narratives:Understanding word-of-mouth marketing in online communities. Journal ofMarketing, 74, 7189.

Lang, B. (2010, September 20). Ipsos OTX study: People spend more than half their dayconsuming media. Retrieved from .

Lasch, C. (1991). The true and only heaven: Progress and its critics. New York: Norton.Lenhart, A., & Madden, M. (2007). Social networking websites and teens: An

overview. PEW Internet and American Life Project.Lewicki, R. J., & Bunker, B. (1995). Trust in relationships: A model of trust

development and decline. In B. Bunker & J. Rubin (Eds.), Conict, cooperation andjustice (pp. 133173). San Francisco: Jossey-Bass.

Mangold, W. G., & Faulds, D. J. (2009). Social media: The new hybrid element of thepromotion mix. Business Horizons, 52, 357365.

Marsh, H., Balla, J., & Hau, K. T. (1996). An evaluation of incremental t indices: Aclarication of mathematical and empirical properties. In G. A. Marcoulidesand& R. E. Schmacker (Eds.), Advanced structural equation modeling: Issues andtechniques (pp. 315353). Mahwah, NJ: Erlbaum.

McAlexander, J. H., Schouten, W. John, & Koening, F. H. (2002). Building brandcommunity. Journal of Marketing, 66, 3854.

McAlexander, J. H., & Schouten, J. W. (1998). Brand-fests: Servicescapes for thecultivation of brand equity. In J. F. Sherry (Ed.), Servicescapes: The concept ofplace in contemporary markets (pp. 377402). Chicago: American MarketingAssociation.

Moorman, C., Zaltman, G., & Deshpande, R. (1992). Relationships between providersand users of market research: The dynamics of trust within and betweenorganizations. Journal of Marketing Research, 29, 314328.

Muniz, A., & OGuinn, T. (1996). Brand community and the sociology of brands. Inthe special session, communities of consumption: A central metaphor fordiverse research. In K. P. Corfman & J. G. Lynch (Eds.), Advances in consumerresearch, (Vol. 23, pp. 265266). Provo, UT: Association for Consumer Research.

Muniz, M. A., & OGuinn, C. T. (2001). Brand community. Journal of ConsumerResearch, 27, 412432.

Muniz, A., & Schau, H. J. (2005). Religiosity in the abandoned Apple Newton brandcommunity. Journal of Consumer Research, 31, 737747.

Nunnally, J. C. (1978). Psychometric testing. New York: McGraw Hill.Pavlou, P. A., Liang, H., & Xue, Y. (2007). Understanding and mitigating uncertainty

in online exchange relationships: A principal-agent perspective. MIS Quarterly,31, 105136.

Rheingold, H. (1993). The virtual community. Homesteading on the electronic frontier.MA: Addison-Wesley.

Rheingold, H. (1991). A slice of life in my virtual community. In L. M. Harasim (Ed.),Global networks (pp. 5780). Cambridge, MA: MIT Press.

Ross, C., Orr, E. S., Sisic, M., Arseneault, J. M., Simmering, M. G., & Orr, R. R. (2009).Personality and motivations associated with Facebook use. Computers in HumanBehavior, 25, 578586.

Sarason, S. B. (1974). The psychological sense of community: Prospects for the

n Behavior 28 (2012) 17551767community psychology. San Francisco: Jossey-Bass.Schau, J. H., Muniz, M. A., & Arnould, J. E. (2009). How brand community practices

create value. Journal of Marketing, 73(5), 3051.

Schembri, S., Merrilees, B., & Kristiansen, S. (2010). Brand consumption andnarrative of the self. Psychology and Marketing, 27, 623637.

Shah, D., Holbert, L., & Kwak, N. (1999). Expanding the virtuous circle of socialcapital: Civic engagement, contentment and interpersonal trust. Paper presentedat annual meeting of Association for Education in Journalism and MassCommunication. New Orleans.

Sherry, J. F. Jr., (1983). Gift giving in anthropological perspective. Journal ofConsumer Research, 10, 157168.

Srinivasan, S. S., Anderson, R., & Ponnavolu, K. (2002). Customer loyalty in e-commerce: An exploration of its antecedents and consequences. Journal ofRetailing, 78, 4150.

Steenkamp, J. B. E. M., & Geyskens, I. (2006). How country characteristics affect theperceived value of a website. Journal of Marketing, 70, 136150.

Szmigin, I., & Reppel, A. E. (2001). Internet community bonding: The Case ofmacnews.de. European Journal of Marketing, 38, 626640.

Tardini, S., & Cantoni, L. A. (2005). A semiotic approach to online communities:Belonging, interest and identity in websites and video games communities. In:Proceedings of IADIS international conference (pp. 371378). Qawra, Malta.

Turkle, S. (1996). Virtuality and its discontents: Searching for community incyberspace. American Prospect, 24(4), 5057.

Von Hippel, E. (2005). Democratizing innovation. Boston, MA: MIT Press.Walden, E. (2000). Some value propositions of online communities. Electronic

Markets, 10, 244249.Walther, J. B. (1996). Computer-mediated communication: Impersonal, inter-

personal and hyper-personal interaction. Communication Research, 23(1), 343.Wang, Y. D., & Emurian, H. H. (2005). An overview of online trust: Concepts,

elements, and implications. Computers in Human Behavior, 21, 105125.

Ward, J. C., & Ostrom, A. L. (2006). Complaining to the masses: The role of protestframing in customer-created complaint web sites. Journal of Consumer Research,33, 220230.

Weber, M. (1978). Economy and society. Berkeley, CA: University of California Press[1922].

Webster, F. E. Jr., (1992). The changing role of marketing in the corporation. Journalof Marketing, 56(October), 117.

Wellman, B., & Gulia, M. (1999). Net-surfers dont ride alone: Virtual communitiesas communities. In B. Wellman (Ed.), Networks in the global village(pp. 331366). Boulder, CO: Westview.

Wellman, B. (1997). An electronic group is virtually a social network. In B. Wellman(Ed.), Culture of the internet (pp. 179205). Mahwah: Lawrence Erlbaum.

Wellman, B. (1999). The network community: An introduction to networks in theglobal village. In B. Wellman (Ed.), Networks in the global village (pp. 148).Boulder, CO: Westview.

Werry, C. (1999). Imagined electronic community: Representations of virtualcommunity in contemporary business discourse. First Monday, 4(9).

Wilk, R. (1996). Learning to not-want things. Paper presented at the association forconsumer research annual conference.

Williams, L., & Cothrell, J. (2000). Four smart ways to run online communities. SloanManagement Review, 41, 8191.

Williams, T., & Williams, R. (2008). Adopting social media: Are we leaders,managers or followers? Communication World, 25(4), 3437.

Wolin, J. S., & Bennett, L. (1984). Family rituals. Family Process, 23(3), 401420.Zhou, Z., Zhang, Q., Su, C., & Zhou, N. (2011). How do brand communities generate

brand relationships? Intermediate mechanisms. Journal of Business Research.http://dx.doi.org/10.1016/j.jbusres.2011.06.034.

M. Laroche et al. / Computers in Human Behavior 28 (2012) 17551767 1767

The effects of social media based brand communities on brand community markers, value creation practices, brand trust and brand loyalty1 Introduction2 Theoretical background2.1 Social media2.2 Brand community2.3 Social media based brand communities

3 Development of the model and the hypotheses3.1 Effects of social media based brand community on community communalities3.1.1 Shared consciousness of kind3.1.2 Shared rituals and traditions3.1.3 Moral responsibilities/obligations to society

3.2 Value creation practices3.2.1 Social networking practices3.2.2 Community engagement practices3.2.3 Impression management practices3.2.4 Brand use practices

3.3 Effects of value creation practices on brand trust and brand loyalty

4 Method4.1 Subjects and procedure4.2 Measures4.2.1 Brand community4.2.2 Shared consciousness, shared rituals and traditions, and obligations to society4.2.3 Social networking4.2.4 Community engagement4.2.5 Brand use4.2.6 Impression management4.2.7 Brand trust4.2.8 Brand loyalty

5 Results5.1 Internal consistency5.2 Unidimensionality5.3 CFA results5.4 Discriminant validity5.5 Convergent validity5.6 Structural model estimation

6 Discussion and implications7 Limitations and future research8 ConclusionAcknowledgementsAppendix A Summary of measuresReferences