Keng2011.pdf

The current issue and full text archive of this journal is available at www.emeraldinsight.com/1066-2243.htm

INTR 21,4

408

Effects of virtual-experience combinations on consumer-related “sense of virtual community” Ching-Jui Keng, Hui-Ying Ting and Ya-Ting Chen

Received 27 September 2010 Revised 4 April 2011 Accepted 26 April 2011

National Taipei University of Technology, Taipei, Taiwan Abstract Purpose – The aim of this study is to investigate the effects of different virtual product experiences (VPEs, interpersonal and machine) on sense of virtual community (SOVC) for consumers. Furthermore, parasocial interaction (PSI) and the sequence of VPEs likely moderate the relationship between different combinations of VPEs and SOVC. Design/methodology/approach – The study employed a factorial online experimental design: 8 (combinations of VPEs, between subjects) x 2 (sequences of VPEs, between subjects). Participants were randomly assigned to the 16 groups. Findings – The results show that with a high degree of PSI, the “escapism þ social” VPEs group has the greatest beneficial effects on SOVC, whereas with a low degree of PSI, the “aesthetics þ information” VPEs group has the greatest beneficial effects on SOVC. Besides, in VPES sequence 1 (interpersonal ! machine), the “escapism þ social” VPEs group has a greater effect on SOVC, whereas in sequence 2 (machine ! interpersonal), there was no statistically significant difference for VPEs combinations relative to SOVC. Practical implication – This study determines that consumers with different degrees of PSI differ from one another regarding their interactions with web sites and communities. Therefore, the characteristics of PSI provide manufacturers with various marketing strategies for interaction. The combinations and sequences of interpersonal and machine VPEs in this study can provide manufacturers with suggestions for appropriate online shopping web site design. Originality/value – Although the importance of these virtual experiences is continuously increasing, there is still a lack of studies that empirically analyze from the combinations of machine interaction and interpersonal interaction of VPEs of the effects on consumer behavior. Keywords Virtual product experience, Parasocial interaction, Sense of virtual community, Virtual worlds, Consumers, Internet Paper type Research paper

Introduction The web is not only a simulation of a real-world environment but also an alternative to a real-world environment. It also changes the ways to companies interacting with consumers (Ozuem et al., 2008). Consumers are able to interact with products in virtual environment and simulating a new form of experience – a virtual product experience Internet Research Vol. 21 No. 4, 2011 pp. 408-434 q Emerald Group Publishing Limited 1066-2243 DOI 10.1108/10662241111158308

The authors thank Dr Chi-Shan Chiang and Dr Tze-Hsien Liao for their constructive comments and feedback on this paper. The authors would also like to thank the Editor, anonymous reviewers’ comments and the National Science Council of the Republic of China, Taiwan for financially supporting this research under contract No. NSC 94-2416-H-027-003.

(VPEs) (Daugherty et al., 2008). According to Hoffman and Novak (1996), VPEs stem from an interaction between consumers and web sites or among consumers through the Internet. Interactivity in computer-mediated environments (CMEs) involves interpersonal interaction and machine interaction. Internet-based, interactive marketing enables consumers and firms to access hypermedia content (machine interaction), and to communicate through the medium (interpersonal interaction). Past-related studies (e.g. Klein, 1998; Li et al., 2001, 2002; Griffith and Chen, 2004; Daugherty et al., 2008) examined chiefly the advertising effects of VPEs from machine interaction (e.g. 3-D advertising). Although the importance of these virtual experiences is continuously increasing, there is still a lack of studies that empirically analyze from the VPEs combinations of machine interaction and interpersonal interaction. With respect to virtual experiences, Thorbjornsen et al. (2002) suggest that on personalized web sites, users with rich Internet experiences have developed strong relationships with a brand, whereas on community web sites, users with rich Internet experiences develop weak relationships with a brand. Koh and Kim (2003) propose that sense of virtual community (SOVC) is affected by community characteristics and affects the strength of the relationship between a consumer and a product. The members’ perception of community characteristics including perceived of usefulness, perceived ease of use and offline activities have positive effects on sense of belonging to virtual communities (Lin, 2007). Therefore, combining different interpersonal and machine VPEs may significantly influence consumer SOVC. This study investigates the effects of different virtual experiences (interpersonal and machine) on the brand’s SOVC for consumers. Li et al. (2002) applied an experimental design to verify the effects of VPEs in different sequences, and demonstrate that virtual experience presented before indirect experience or direct experience produces the best learning effect. Marks and Kamins (1988) also propose that the indirect experience presented before direct experience has the best persuasive effect. The current study treats sequences of different virtual experiences as a moderating variable to examine whether different sequences of VPEs affect consumers. Horton and Wohl (1956) introduced the notion of parasocial interaction (PSI), an illusion-based intimate relationship between an audience and media. Audiences may be engaged in PSI owing to their dependence on, for example, home-shopping channels, and may see such channels as sources of credible information (Rubin and Step, 2000). Ballantine and Martin (2005) propose that increased PSI likely influences access to online communities. The current study also uses the high or low degrees of PSI as a moderating variable in determining whether different combinations of virtual experiences and PSI may result in different SOVCs. Virtual product experiences (VPEs) Consumers learn from experience, that is, they learn by interacting with environments (Li et al., 2001). On the basis of different degrees of interaction, such experiences can be classified as Direct Product Experience (DPE) and Indirect Product Experience (IPE). A DPE is a phenomenon during which information is acquired by a consumer who directly interacts with a product, in contrast to IPE, during which a consumer’s experience with a product is mediated, perhaps through traditional advertising. Notably, DPE increases consumer expectations regarding future purchases (Chiou et al., 2008). Klein (1998) demonstrates that the medium is a dominant source of

Effects of virtual experiences

409

INTR 21,4

410

information, through which consumers can virtually experience product functions. The Virtual Product Experiences (VPEs) is the same concept with telepresence. Steuer (1992) defined “telepresence” as the mediated perception of an environment via the Internet. Telepresence occurs when customers focus on an Internet environment and feel that the virtual environment with which they are interacting is both more real than and dominates the actual physical environment. Fundamentally, VPEs is by means of a product representation that activates consumer mental processes via a sensory-rich mediated environment (Ozok and Komlodi, 2009; Daugherty et al., 2008; Fortin and Dholakia, 2005). Burke (1997), who set up a virtual store using 3-D computer animation, proposes that when the richness of VPEs increases, the persuasiveness of advertising increases. Notably, VPEs is similar to IPE in that both are mediated experiences, also resembling DPE as both are interactive (Hoch and Deighton, 1989). VPEs derive from a consumer’s personal experience, which has various interaction modes. Interactivity in a CME has interpersonal interaction and machine interaction (Hoffman and Novak, 1996). Interpersonal interaction includes dialogue, communication, and the exchange of information among users via a medium that creates interpersonal virtual experience, including chat room, MSN, virtual community. An online community is a mode that produces interpersonal interaction. Machine interaction is the interaction between users and content via hypermedia that create machine virtual experience. A commercial web site is a mode that produces machine interaction. Interpersonal VPEs Virtual communities include people who share similar interests, experiences, expertise, or backgrounds. The community web site also offers an environment in which users can socialize and interact with other members and acquire social experiences. In fact, both community experiences based on similar backgrounds or interests and exchanges of information with other members, which may comprise virtual experiences, influence consumers’ learning and brand relationships (Mandel and Johnson, 2002). Armstrong and Hagel (1996) argue that interactions in virtual communities are based on four basic needs: transaction, interest, fantasy, and relationship. The interactions among virtual-community members are also sorted into four categories: (1) information; (2) relationship; (3) entertainment; and (4) conversation. Thorbjornsen et al. (2002) compared the ability to develop consumer-brand relationships between personalized web sites and customer communities, used two different products – airline ticker and restaurant meal – to strengthening external validity. According the conclusion of Thorbjornsen et al. research, customer communities developed stronger relationships among respondents with limited Internet experience than among respondents with extensive Internet experience. Voorveld Neijens and Smit (2009) provide an integrated literature review of factors influencing consumers’ response to brand community web sites. The review paper shows that the perceived interactivity of the community web site positively influenced

affective response to the community. When consumers perceived the members as more interactive, they also had a more positive impression to the community. Zhou (2011) examined the determinants of online community user participation from a social influence perspective, based on 450 valid responses collected from a survey questionnaire, the results show that both social identity and group norm have significant effects on user participation. Koh and Kim (2003) developed a measurement index for SOVC based on the community psychology literature and on interviews with the leaders of diverse virtual communities in Korea. In Koh and Kim’ research, a list of 50 virtual communities willing to participate in the study was obtained from the Internet community service providers, 44 of these communities were choice because they satisfy the selection criteria (i.e. they have at least 20 members and have been in operation for at least two months). They propose that SOVC is affected by community features, namely, enjoyability, leaders’ enthusiasm, and offline activities, which engender different relationship levels between consumers and products. Among these characteristics, enjoyability significantly impacts consumer SOVC. Also, enjoyability resulting from interaction may produce a flow effect and enhance community loyalty. Wooldridge (1999) proposes that consumers, while receiving a service, may have subjective experiences from interpersonal interactivities therein. According to the levels of intensity and intimacy in interactions among community members, interpersonal VPEs have two forms of virtual experience: (1) social VPEs—with high degrees of intimacy and interaction—rests on consumers’ interpersonal interactions, such as discussions, explanations, and online polls; and (2) information VPEs, which generates low levels of intimacy and interaction, and which rests on infrequent interpersonal interaction activities such as browsing online articles and responding to specific articles. Machine VPEs Pine and Gilmore (1999) used two dimensions to describe experiences. One dimension refers to the involvement of individuals with one end of an axis representing passive participation, indicating that consumers do not directly influence performance; the other end of the axis represents active participation, indicating that a consumer acts in the chief role in sharing and creating related events. The other dimension assesses the connection type, with one end of the axis representing absorption, which is a way to attract attention via firsthand information; the other end of the axis represents immersion, indicating that a consumer becomes part of a real-life environment. The current study uses telepresence (Steuer, 1992) and the persuasion routes of the Elaboration Likelihood Model (ELM) (Petty et al., 1983) to develop four forms of machine VPEs. Viewing an individual’s presence in the media as a telepresence is appropriate, and stands in contrast to the involvement dimension of Pine and Gilmore (1999). Machine interactivity refers to an interaction between a person and a CME. Steuer (1992) used the term “telepresence” to describe created mediated environments and claimed that when consumers interact with a CME, the consumers adopt the perceptions common to the physical environment in which these consumers are present. We can employ vividness and interactivity of telepresence, concluding that a high degree of telepresence is related to high degrees of vividness and interactivity. Passive participation is associated with a low degree of telepresence, whereas active

Effects of virtual experiences

411

INTR 21,4

412

Figure 1. The relationship among ELM, consumer learning and persuasion

participation is associated with a high degree of telepresence. However, Cho (1999) used the ELM to describe the process of consumers’ interacting with advertisement information on web sites. In the ELM model, consumers are persuaded by the central route of high elaboration likelihood when browsing online advertisements with a high degree of involvement, motivation, and ability to obtain desired knowledge and information. Comparatively, consumers with low elaboration likelihood may be unwilling or unable to devote energy; in this situation, persuasive factors, such as appealing pictures, text, and music, influence persuasion effects, namely, along the peripheral route. Figure 1 shows the linkage among ELM, consumer learning (IDE, VPE, and DPE) and persuasion. In this study, the four forms of machine VPEs are illustrated in Figure 2 and are as follows: (1) Education VPEs is the experience consumers obtain directly from manufacturer-provided product information and has a low degree of telepresence (low in vividness and middle in interactivity); thus, consumers may be persuaded by the consumer-based central route. (2) Entertainment VPEs is the experience surrounding consumer-acquired information provided by a manufacturer via peripheral entertainment; this experience is not directly involved in an activity that has a high degree of telepresence (high in both vividness and interactivity). (3) Aesthetics VPEs may attract consumers by aesthetics not directly related to an activity, and this has a low degree of telepresence (high in vividness and low in interactivity).

Effects of virtual experiences

413

Figure 2. Machine VPE in this study

(4) Escapism VPEs is the experience generated by consumers completely immersed in manufacturer-provided product information and is directly related to an activity and, thus, falls into the category of a high degree of telepresence (high in vividness and interactivity). According to the current study’s classification of machine VPEs, machine VPEs offer entertainment, aesthetics, education, and escapism. Interpersonal VPEs, according to the virtual-community classifications of Armstrong and Hagel (1996), involves social VPEs and information VPEs. Interpersonal VPEs (two types) and machine VPEs (four types) produce eight different groups. Sense of virtual community (SOVC) McMillan and Chavis (1986) defined SOC as a sense of belonging, consideration for other group members and the entire group, sharing and trust, and meeting needs and commitments. Blanchard and Markus (2004) defined SOC as having the following four features: (1) Feelings of membership: a sense of belonging to, and identifying with a community. (2) Feelings of influence: a sense of having influence on and being influenced by a community. (3) Fulfillment of needs: a sense of being supported by others in a community while also supporting others. (4) Feeling of sharing: a sense of relationships, shared experience, and community spirit. Koh and Kim (2003) coined the term “SOVC” to define the term “SOC” as it applies to virtual environments; used these dimensions to define SOVC as consisting of membership, influence, and immersion. The characteristics of a virtual community, including enjoyability, leader enthusiasm, and offline activities will influence the SOVC. Enjoyability can be derived from the interaction among members or between

INTR 21,4

414

members and text. When a member’s needs are fulfilled via interaction, enjoyability is produced and exerts a positive effect on SOVC. Thus, enjoyability directly relates to virtual experiences that engender the same effects as do offline activities. Barnes and Pressey (2011) also propose that arousal and pleasure are potent drivers of higher-order needs (belonging) in participation of virtual community. Li et al. (2002) employed an experimental design to verify the effects of various sequences of virtual experiences (direct and indirect); however, their study simply replaces virtual experiences with 3-D computer animation without investigating the different combinations of virtual experiences. Furthermore, Koh and Kim (2003) argue that community enjoyability affects membership and immersion, and enjoyability is directly related to VPEs. Consequently, we assume that different combinations of interpersonal VPEs (community web site) and machine VPEs (3-D computer animation and advertising) may engender different SOVC. According to the four forms of machine VPEs in current study, Entertainment VPEs surrounds the main information provided by a web site via peripheral entertainment; that has a high degree of telepresence (high in both vividness and interactivity). Escapism VPEs are experiences that derive from consumers and that fall completely into the category of a high degree of telepresence (high in both vividness and interactivity). Hence, the two types of experiences that consumers generate are immersed completely in web site-based information. Furthermore, social interpersonal VPEs involve interpersonal interactions, with an emphasis on the degree to which participants in a communication process hold similar values (Nicholson et al., 2001). Zhou (2011) also found that both social identity and group norm that produced by social interpersonal VPEs have significant effects on user participation of virtual community. Hence, social VPEs produce a greater sense of belonging to and identifying with a community than information VPEs. Consequently, we assume that group 4 (Entertainment þ Social) and group 8 (Escapism þ Social) have the greatest effects on SOVC. H1a.

Different combinations of VPEs result in different SOVC for consumers.

H1b.

The combinations of group 4 (Entertainment þ Social) and group 8 (Escapism þ Social) have the greatest effects on SOVC for consumers.

Moderating effects of the sequence of VPEs Marks and Kamins (1988) propose that the sequence of product commercials (indirect experience) presented before product trials (direct experience) has the best persuasive effect. Millar and Millar (1996) note that direct experiences generate more affective reactions than indirect experiences; however, indirect experiences generate more cognitive reactions than do direct experiences. Suh and Lee (2005) investigated the influence of virtual experience on consumer acquisition of product knowledge and argue that consumers with strong virtual experiences can learn more effectively than those with weak virtual experiences. In their research, strong virtual experience resembles machine VPEs as defined in the current study, whereas weak virtual experience resembles interpersonal VPEs in the current study. Li et al. (2002) used an experimental design to verify the effects of virtual experience in different sequences, and demonstrate that virtual experience presented before indirect experience and virtual experience presented before direct experience produce the best learning effect. Nevertheless, their study replaces VPEs with 3-D computer animation only, short of

classifying and combining different types of virtual experiences. In fact, a virtual community can generate virtual experience and positively affect consumer learning (Mandel and Johnson, 2002). In the virtual environment in which interpersonal interactivity is positioned before machine interactivity, web sites with a high degree of social interaction serve to facilitate experience sharing and communication among consumers and to produce a sense of belonging. Herein, the consumer’s elaboration will increase. Therefore, when machine VPEs is coupled with a high degree of telepresence (high vividness and high interactivity (escapism VPEs)), these web sites lead consumers to obtain more elaboration and is further persuaded through the central route of ELM (Petty et al., 1983). Then, these consumers will produce more SOVCs. In such a case, combining “escapism þ social” experiences has the best effects. When machine VPEs comes before interpersonal VPEs, the peripheral cue with high telepresence (high vividness and high interactivity (entertainment VPEs)) web sites serves to garner consumer attention. Then, high social interaction community web sites aid consumers the production of a sense of belonging via experience sharing and communication. Accordingly, combining “entertainment þ social” experiences has the best effects. We propose the following hypotheses: H2a.

For the sequence of “interpersonal ! machine” VPEs, the combination of Group 8 (Escapism þ Social) experiences has the best effects on SOVC for consumers.

H2b.

For the sequence of “machine ! interpersonal” VPEs, the combination of Group 4 (Entertainment þ Social) experiences has the best effects on SOVC for consumers.

Moderating effects of parasocial interaction (PSI) Horton and Wohl (1956) developed the PSI paradigm to describe one-sided, parasocial relationships between the mass media and audiences. They and their adherents have proposed that users treat program moderators as good friends; this intimacy builds trust in and reliance on the medium. Rubin and Step (2000) note that many PSI characteristics are conceptually similar to those of social interaction and that, thus, PSI can be interpreted as social interaction or interpersonal relationships. Hoerner (1999) suggests that designers of some commercial web sites have developed virtual representations as a method of attracting the attention of those browsing the web site, which can decrease user uncertainty toward an environment. Hoerner (1999) modified the PSI scale (Rubin et al., 1985) to generate the PSI-Web scale, which serves to assess the degrees of interpersonal interaction created during Internet surfing. Hoerner (1999) analyzed four commercial web sites, concludes that interpersonal interaction benefits users and increases user willingness to visit the web site again. Stephens et al. (1996) used a QVC case study to show that hosts systematically form parasocial interactions to encourage viewers to purchase products from QVC. Rubin and Step (2000) demonstrates that PSI between a TV host and an audience may lead the audience to treat the host as a credible source of information; that is, face-to-face interpersonal interaction is replaced by interaction with the medium. However, most studies focus on PSI derived from traditional media and few have investigated PSI derived from web

Effects of virtual experiences

415

INTR 21,4

416

sites. This study analyzes the potential moderating effects of degrees of PSI for consumers in virtual communities. Ballantine and Martin (2005) argue that PSI positively influences community members and that a high level of member participation strengthens the commitment of other users to the communities. Thus, users with a high degree of PSI may be admired and depended upon and may become loyal to the information source (Rubin and Step, 2000). Thus, consumers with different degrees of PSI have different demands for interaction with different web sites and communities. That is, under virtual experience combined with different degrees of interactivity and vividness, people with high degrees of PSI are highly active and tend to respond to persuasions via central routes. Therefore, in treating people with high degrees of PSI, a highly interactive social virtual community can encourage consumers to share experiences, communicate and produce a sense of belonging. Escapism machine VPEs has a high degree of telepresence (high vividness and high interactivity) in inducing consumers to obtain entertainment and immersion effect. Keng and Ting (2009) also found high PSI would lead to a positive relation with entertainment value of web blogs. Hence, combining “escapism þ social” VPEs has a positive effect on SOVC. Conversely, consumers with low degrees of PSI are typically passive and tend to be persuaded by the peripheral route of ELM (Petty et al., 1983); thus, the current study supposes that when the marketer is dealing with people possessing a low degree of PSI and telepresence (high vivid and low interactivity), aesthetic web sites should attract consumer attention, causing the web sites’ visitors to exhibit evidence of the immersion effect. The information VPEs with a low degree of interactivity and intimacy generate a sense of belonging by communicating and inquiry infrequently such as browsing online articles. Thus, combining “aesthetics þ information” VPEs has a beneficial effect on SOVC, as the following hypotheses state: H3a.

The combination of Group 8 (Escapism þ Social) VPEs has the best positive effects on SOVC with a high degree of PSI.

H3b.

The combination of Group 5 (Aesthetics þ information) VPEs has the best positive effects on SOVC with a low degree of PSI.

Method This research analyzes the effects of each of the eight VPEs combinations on SOVC. Furthermore, this study treats the sequence of interpersonal and machine VPEs and PSI as moderating variables between experience combinations and SOVC (Figure 3). Participants The current study employed a factorial design: 8 (VPEs, between subjects) x 2 (sequence of VPEs, between subjects). Participants were randomly assigned to the 16 groups. This study used an online survey. In total, the study collected 937 questionnaires, of which 791 were valid (Table I). According the survey report in 2010 of Internet usage in Taiwan by TWNIC (Taiwan Network Information Center), the percentage of male of Internet users is 74.16 percent, and the female is 70.93 percent. The percentages in different age groups of Internet user who have Internet experiences in Taiwan, age 20-24 have the highest percentages, 95.11 percent; the second is the group of age 15-19 with 93.92 percent; the next is the group of age 25-34 with 91.35

Effects of virtual experiences

417

Figure 3. The conceptual model for this research

Characteristics

Item

Sex

Female Male 0-14 15-28 29-42 Over 42 Junior high school Senior high school College Graduate school 0-1 2-3 4-5 Over 5 No Yes

Age

Education

Internet-use experience (years)

“Community web site”-use experience (Yes/No)

Frequency

(%)

114 677 17 544 200 30 29 153 510 99 19 37 160 575 54 737

14.4 85.6 2.1 68.8 25.3 3.8 3.7 19.3 64.5 12.5 2.4 4.7 20.2 72.7 6.8 93.2

Note: N ¼ 791

percent. Among Internet users of age 12 and above with experience of using the Internet, 48.09 percent of them have used more than 10 years, 29.92 percent with over 5 years and less ten years. According the educational background, 83.63 percent of Internet users have junior college degree. There is consistency between this finding and the current study in age, education and Internet-use experience, but there are differences on sex. In accordance with the survey reports of consumer behavior of Internet in Taiwan by MIC (Market Intelligence & Consulting Institute), female prefer buying dresses and cosmetics, male prefer the 3C products. The digital camera was chose as the experimental product for designing the experimental web sites of VPEs, consequently the percentage of male exceeding female in the sample. In addition, we examined the significance between these characteristics and SOVC by MANOVA, the analytical results revealed there is not significant differences between all

Table I. Main characteristics of the sample

INTR 21,4

characteristics and SOVC (Sex p ¼ 0.308; Age p ¼ 0.142; Education p ¼ 0.247; Internet – use experience p ¼ 0.913; “Community web site” use experience p ¼ 0:361Þ: However, we used a Pearson’s chi-square test for cross tabulation to analyze the randomization of the main characteristics of the sample in 16 groups, the result indicating these characteristics did not differ from one another significantly in each group, hence, can reduce the influence of the proportion difference of sex.

418 Interpersonal VPEs design According to degrees of interpersonal interaction and intimacy, this study divided interpersonal VPEs into two type of community (social VPEs and information VPEs). The information contents about the digital camera are consistent in social community and information community. In Information VPEs (low interpersonal interaction) emphasize manufacturer messages, test reports and related news about the Fuji Z2 digital camera. Only certain people post articles. Social VPEs (high interpersonal interaction) emphasize feedback and sharing after use of digital camera; also, in the context of the VPEs, there is greater diversity among people who posted and responded to articles (Table II). Machine VPEs design Stimuli are distinguishable by the differences associated with each experience. The direct experience allowed the full sensory inspection of the real product and print advertising provided the indirect experience (Daugherty et al., 2008). The VPEs is designed after a literature review (Keng and Lin, 2006; Hoffman and Novak, 1996; Petty et al., 1983; Steuer, 1992). This study adopted vividness (Steuer, 1992) and interactivity (Keng and Lin, 2006) to design situations in which experimental web sites would provide the following four realms of VPEs: (1) Entertainment VPEs (high telepresence, peripheral route) – emphasize the interesting by 2-D Flash web pages, interesting images and text, online games. In addition, the characteristic of Fuji Z2 digital camera appears with Flash web page. (2) Aesthetic VPEs (low telepresence, peripheral route) – present the product information about Fuji Z2 digital camera on Flash web pages and in 2-D dynamic images of high quality. Also, emphasize the abundance of background music, text colors and button colors.

Types of community Information

Social Table II. The design of interpersonal VPEs

Contents of community

Interpersonal interaction

Chiefly the manufacturer messages, test reports and related news about the digital camera. Only certain people posted articles Chiefly feedback and sharing after using the digital camera; greater diversity among people who posted and responded to articles

Low Involving fewer communications and inquiries with lower response rates High Involving more communications and inquiries with higher response rates

(3) Education VPEs (low telepresence, central route) – emphasize the learning of digital camera product information. Also, use static images and text to specify general product information about the Fuji Z2 digital camera. Increase the interactivity by the way to link. (4) Escapist VPEs (high telepresence, central route) – show the product information about the Fuji Z2 digital camera appears on Flash web pages. Offering the 3-D simulation situation of Fuji camera, so that, consumers could experience the effect of trying the digital camera (Table III).

Effects of virtual experiences

419

Pretest Experience products are those that consumers want to try in order to reduce their uncertainty, before making a purchase (Nelson, 1970, 1974; Girard and Dion, 2010). Klein (1998) demonstrates that the medium is a dominant source of information, through which consumers can virtually experience product functions. In order to investigate the impact of VPEs combinations, the current study selected experience products on the

VPEs

Content of web sites

Entertainment Emphasizing the interesting by 2-D Flash web pages, interesting images and texts, online games. In addition, the characteristics of the Fuji Z2 digital camera on the Flash web page Aesthetics Showing the product information about the Fuji Z2 digital camera on the Flash web page and 2-D dynamic images of high quality. Emphasizing the abundance of background music, text colors and button colors Education Emphasizing the learning of product information of digital camera. Offering the main specifications and product information about the Fuji Z2 digital camera, in the form of static images and text. Increasing the interactivity by the way to link Escapist Showing the product information about the Fuji Z2 digital camera on the Flash web page. Using the 3-D simulation situation of the Fuji camera, consumers can experience the effects of trying the digital camera

ELM Interactivity Vividness Telepresence route High

High

High

Peripheral

Low

High

Low

Peripheral

Median

Low

Low

Central

High

High

High

Central

Table III. The design of machine VPEs

INTR 21,4

420

basis of Nelson’s (1970) definition. In order to find out the most representative product as an experience product, during the pretest with 59 subjects, the respondents were asked two questions regarding each product from ten product categories: notebook computer, mobile phone, digital camera, contact lenses, clothes, restaurants, wristwatch, shoes, furniture and a tour schedule, which are based on the definition of search/experience product drawn from the literature (Girard et al. 2003; Bei et al., 2004; Franke et al., 2004; Huang et al., 2009). A questionnaire containing 20 items rested on a seven-point Likert-type scale for asked opinions about the ease of evaluating the quality and engaging in information processing of candidate products. By analysis of variance (ANOVA), the results show that scores for different products are significantly different ð p ¼ 0:001Þ: The current study applied a Duncan’s multiple comparison to the ten products and chose the digital camera as the experimental product for designing the experimental web sites of VPEs. Further, to increase the study’s legitimacy and to avoid any situation in which consumer knowledge about a brand would affect the consumers’ inferences, the study selected a reputable digital-camera company and tested preference. Participants were asked to answer two questions about 11 digital-camera brands so that the study could determine the participants’ familiarity with and beliefs about each brand (How do you feel about the quality of brand X’s digital cameras? How familiar are you with brand X’s digital cameras?). In this part, the study chose “Fuji” as the product brand on the basis of the mean score. Dependent variables measurement In this study, dependent variables (PSI and SOVC) were measured on a seven-point Likert-type scale, ranging from (1) strongly disagree to (7) strongly agree. The scale items for PSI were revised according to the PSI-Web scale developed by Hoerner (1999). PSI evaluates the degree to which the users perceived social interactivity on the web site. Items for SOVC were adopted from Koh and Kim (2003) and comprise membership, influence and immersion; however, influence was not affected by enjoyability in the research conducted by Koh and Kim. Hence, the current study did not examine this dimension. Procedure This experiment was conducted on a web site. The web site had a program attached to the web pages to prevent subjects from returning to a previous page or from repeatedly clicking the same web page. Subjects were required to link to the experimental web site, apply for an account and fill out a personal-information form requesting information on their sex, age, educational background, Internet-use experience and “community web site”-use experience. Subjects were then introduced to the experiment and filled out the PSI questionnaire. For the second part, subjects were randomly allocated into one of 16 VPEs groups (Table IV). The third part focused on SOVC evaluation. This experiment reached its final stage when subjects finished the third part of the questionnaire. Subjects then clicked the leave button to close the experimental window. Results Reliability and validity This study assessed reliability by applying CFA. Confirmatory Factor Analysis (CFA) was applied to measure factor loading, individual item reliability and composite

8 VPEs combinations

VPEs sequence

1: Education þ Information

Interpersonal!machine Machine!interpersonal Interpersonal!machine Machine!interpersonal Interpersonal!machine Machine!interpersonal Interpersonal!machine Machine!interpersonal Interpersonal!machine Machine!interpersonal Interpersonal!machine Machine!interpersonal Interpersonal!machine Machine!interpersonal Interpersonal!machine Machine!interpersonal

2: Education þ Social 3: Entertainment þ Information 4: Entertainment þ Social 5: Aesthetics þ Information 6: Aesthetics þ Social 7: Escapism þ Information 8: Escapism þ Social Total

Sample size 49 44 49 50 59 40 42 61 55 47 54 47 55 48 45 46 791

reliability (CR). The current study obtained a recommended value of indicator reliability exceeding 0.5 and its CR reached the ideal value of 0.7 (Fornell and Larcker, 1981), demonstrating the internal consistency of all tested variables (Table V). The validity is also assessed. Drawing on a survey conducted under the principles of relevant research, this study made some adaptations based on the characteristics of the given group surveyed to ensure high content validity. Furthermore, all primary items should have only a factor loading exceeding 0.5, while for other factors, a lower factor loading is best, demonstrating good convergent validity. In another aspect, the correlation estimates between any two factors were very low, and in no case did the confidence intervals contain 1.00, thus indicating that discriminant validity was upheld (Table VI). To ensure internal validity, the study controlled the experimental process on the basis of experimental-web site design, so that each account could yield only one completed questionnaire and so that no subject could repeatedly browse a single web page. The study recorded time taken to answer the questionnaire, time logged in and user IP address to ensure the internal validity of the experiment. For external validity, the study posted the questionnaire on different web sites to increase sampling diversity. Manipulation check Interpersonal VPEs. The degrees of interaction were measured on the scales developed by Srinivasan et al. (2002) and Wu (2002) with modified wording to suit this study. Each item rests on a seven-point Likert-type scale. The study employed an independent sample t-test to examine the degree of interaction between two different interpersonal VPEs (information and social) by 27 subjects participated. The results indicate that the difference in degrees of interaction between the information community and the social community is significant ð p ¼ 0:041 , 0:05Þ (Table VII). Moreover, the degree of interaction of the social virtual community ðmean ¼ 4:92Þ was higher than that of the information virtual community ðmean ¼ 4:38Þ: Thus, the manipulation of interpersonal virtual-experience interaction is successful.

Effects of virtual experiences

421

Table IV. Sample sizes of 16 random groups

INTR 21,4

422

Variables/items Membership for SOVC I feel as if I belong to the virtual community I feel membership in my virtual community I feel as if my virtual community members are my close friends I like my virtual community members Immersion for SOVC I spend much time on-line in my virtual community I spend more time than I expected navigating my virtual community I feel as if I am addicted to my virtual community I have missed classes or works because of my virtual community activities

Table V. Items and result of confirmatory factor analysis

Parasocial interaction This web site adds credibility to the information it provides me I would tell my friends about this web site I feel sorry for this web site when there are mistakes or problems with it This web site is interested in my opinions and comments I feel as if I am part of a close-knit group when I visit this web site Visiting this web site helps me form opinions about the topics and issues presented at this site I would visit this web site again. The personality of this web site is friendly and down-to-earth I can trust the information I get from this web site Visiting this web site made me relax and have fun I wanted to say something to this web site I got mad at this web site when it did not work properly The personality of this web site makes me feel comfortable, as if I am with friends I felt the time I spent visiting this web site was worth it This web site was considerate and didn’t overload my computer’s technical capabilities

Factor loading

t-value

Composite reliability

AVE

0.6730

19.8326

0.8376

0.5640

0.7757

23.8847

0.7787 0.7714

24.0107 23.7126

0.8173

26.4927

0.8494

0.5915

0.8557

28.2678

0.8166

26.4588

0.5467

15.6998

0.7250 0.7492

23.2417 24.3338

0.9392

0.5098

0.5961

18.0356

0.5404

16.0276

0.7066

22.4393

0.7408 0.7813

23.9472 25.8452

0.6411

19.7542

0.7140

22.7610

0.8175 0.7235

27.6592 23.1758

0.7287

23.4075

0.6880

21.6517

0.7828

25.9171

0.7242

23.2061

Note: Goodness-of-fit Index GFI ¼ 0.91 CFI ¼ 0.93 NFI ¼ 0.91 NNFI ¼ 0.93 RMSEA ¼ 0.06

Machine VPEs. The study conducted a manipulation check of machine VPEs by 49 subjects participated based on the literature (Ducoffe, 1996; Pine and Gilmore, 1999; Mathwick et al., 2001; Novak et al., 2000; Barnes and Mattsson, 2008). In order to avoid the experimental error results from the quantity of information content offered by the four different experimental web sites, this study conducted a manipulation check focusing on the quantity of information content first. The scale derived from Ducoffe (1996) who propose that the information content of a web site must be relevant, timely, immediately accessible, and a convenient and good source. An ANOVA test shows that the difference of scores for four experimental web sites was not significantly ðF ¼ 1:646; p ¼ 0:192 . 0:05Þ: Therefore, the quantity of information content provided by the four different experimental web sites was consistent. Furthermore, the education VPEs web site offers the basic information about the Fuji digital camera, emphasizing the learning of product information about digital camera, so views the education experimental web site as the control group in order to check whether or not the difference of the entertainment experience, aesthetics experience and escapism experience provided by four experimental web sites is significant. The entertainment experience was measured by using the scale developed by Ducoffe (1996), who argues that an entertainment experience must be entertaining, enjoyable, pleasing, fun and exciting. An ANOVA test shows that scores for the four experimental web sites significantly differed from one another ðF ¼ 5:135; p ¼ 0:04 , 0:05Þ; the mean (26.00) for the entertainment VPEs web site was highest, and the entertainment-experience difference between the entertainment VPEs web site and the education VPEs web site is significant ð p ¼ 0:032 , 0:05Þ: Pine and Gilmore (1999) suggest that the color, space and art used in a design attract users and produces an aesthetic visual experience. In the current study, the aesthetic experience was measured by using the scale developed by Mathwick et al. (2001). The ANOVA test shows that scores for the four experimental web sites significantly differed from one another ðF ¼ 3:348; p ¼ 0:027 , 0:05Þ; the mean (21.25) for the aesthetic VPEs web site was highest, and the aesthetic-experience difference between the aesthetic VPEs web site and the education VPEs web site is significant ð p ¼ 0:033 , 0:05Þ: Barnes and Mattsson (2008) argue

Membership Immersion PSI

Membership

Immersion

PSI

0.75 0.21 [0.13, 029] 0.04 [0.04, 012]

0.77 0.022 [0.15, .30]

0.71

Notes: Square root of AVE in italics on the diagonal; [] indicates confidence intervals of the correlation estimates between any two factors

Information VPEs Interpersonal VPEs Note: *p , 0.05

Size

Mean

Std.

t-value

p-value

14 13

4.3810 4.9231

0.5040 0.7836

2 2.155

0.041 *

Effects of virtual experiences

423

Table VI. Discriminant validity

Table VII. T-test value (the degree of interaction between two different interpersonal VPEs)

INTR 21,4

424

that escapism experiences stemming from such mechanisms as a 3-D virtual environment have high telepresence. The escapism experience was measured by using the scale developed by Novak et al. (2000). The ANOVA test shows that scores for the four experimental web sites significantly differed from one another ðF ¼ 4:702; p ¼ 0:006 , 0:05Þ; the mean (32.17) of the escapism VPEs web site was highest and the escapism-experience difference between the escapism VPEs web site and the education VPEs web site is significant ð p ¼ 0:029 , 0:05Þ. Thus, the manipulation of the machine VPEs experimental web sites was successful. The following are the items for manipulation check: (1) Interpersonal VPEs . I feel the degree of interpersonal interaction is very high in this community; . I feel the connection between each member is very strong in this community; . I feel the interaction between each member is true to actual life in this community. (2) Machine VPEs . Quantity of information content: – the web site is a good source of product information about the Fuji Z2 digital camera; – the web site supplies relevant product information about the Fuji Z2 digital camera – the web site provides timely product information about the Fuji Z2 digital camera; – the web site is a good source of up-to-date product information about the Fuji Z2 digital camera – the web site makes product information about the Fuji Z2 digital camera immediately accessible. . Entertainment experience: – I find it entertaining to browse the Fuji Z2 digital camera web site; – I enjoy browsing the Fuji Z2 digital camera web site; – I feel that the Fuji Z2 digital camera web site is very pleasant; – It is fun to use the Fuji Z2 digital camera web site; – It is exciting for me that browsing the Fuji Z2 digital camera web site. . Aesthetics experience: – the way the Fuji Z2 digital camera web site displays its products is attractive; – I like the way the Fuji Z2 digital camera web site looks; – I feel that the Fuji Z2 digital camera web site is aesthetically appealing. . Escapist experience: – I forget about my immediate surroundings when I browse the Fuji Z2 digital camera web site; – using the Fuji Z2 digital camera web site often makes me forget where I am; – after using the Fuji Z2 digital camera web site, I feel like I have come back to the “real world” after a journal; – using the Fuji Z2 digital camera web site creates a new world for me, and this world suddenly disappears when I stop browsing the site;

– when I use the Fuji Z2 digital camera web site, I feel I am in a world created by the web site I visit; – when I use the Fuji Z2 digital camera web site, my body is in the room, but my mind is inside the world created by the web site I visit; – when I browse the Fuji Z2 digital camera web site, the world generated by the web site I visit is more real for me than the “real world”.

Effects of virtual experiences

425 Randomization Subjects were randomly divided into 16 groups after providing information about the five factors: (1) sex; (2) age; (3) educational background; (4) Internet-use experience; and (5) “community web site”-use experience. We used a Pearson’s chi-square test for cross tabulation to analyze the randomization of the five factors in each group. The p-value of each factor is not significant, indicating that in each group, the five factors did not differ from one another significantly in this regard. Thus, the manipulation of randomization was successful. Hypothesis testing. According to machine virtual experience and interpersonal virtual experience, subjects were divided into eight combinations. The effect of different VPEs groups on SOVC (membership and immersion) was tested by means of a one-way MANOVA in the current study. Results of the MANOVA indicate significant main effects of VPEs combinations (Wilks’s l ¼ 0.959, F ¼ 2.358, p ¼ 0.003 , 0.05). Since the main effect of eight different VPEs combinations is significant, the means of the cells are examined by performing LSD multiple comparisons. As shown in Table VIII, in the immersion of SOVC, the effect of group 4 (Entertainment þ Social) was significantly better than the effect of group 1 (Education þ Information), group 2 (Education þ Social) and group 6 (Aesthetics þ Social) ð p ¼ 0:0040:0430:019 , 0:05Þ; the effect of group 8 (Escapist þ Social) was significantly better than only the effect of group 1 (Education þ Information) (p ¼ 0.019 , 0.05) (Figure 4). In the membership of SOVC, the effect of group 4 (Entertainment þ Social) was not better than other groups; the effect of group 8 (Escapist þ Social) was significantly better than only group 4 (Entertainment þ Social) and group 5 (Aesthetics þ Information) ð p ¼ 0:014 0.018 , 0.05) (Figure 5). Consequently, H1a was confirmed, H1b was partial accepted. The analyses rested on a division of the subjects into two groups according to the VPEs sequence: interpersonal VPEs is first; and machine VPEs is first. Referring to the sequence of experiences and the eight experience combinations, this study applied a two-way MANOVA to SOVC (membership and immersion). The analytical results reveal that the interaction effect was not significant between the experience sequence and the experience combinations ðWilks’sl ¼ 0:987; F ¼ 0:701; p ¼ 0:775 . 0:05Þ: Further, we tested the main effect in different sequences of VPEs by the one-way MANOVA (Figure 6 and Figure 7). In sequence 1 (interpersonal ! machine), eight combinations of VPEs resulted in different SOVC (Wilks’s l ¼ 0.942, F ¼ 1.715,

INTR 21,4

SOVC Immersion

Group I

Group J

Mean (I – J)

Standard error

p 2 value

4

1 2 3 5 6 7 8 1 2 3 4 5 6 7 1 2 3 5 6 7 8 1 2 3 4 5 6 7

0.3266 * 0.2227 * 0.0485 0.2036 0.2577 * 0.1189 0.0565 0.2700 * 0.1662 2 0.0081 2 0.0565 0.1470 0.2011 0.0624 2 0.0829 2 0.2438 * 2 0.1150 2 0.0095 2 0.2346 * 2 0.0801 2 0.2407 * 0.1578 2 0.0031 0.1257 0.2407 * 0.2311 * 0.0061 0.1606

0.11179 0.10999 0.10999 0.10916 0.10944 0.10890 0.11243 0.11523 0.11349 0.11349 0.11243 0.11269 0.11295 0.11243 0.09680 0.09525 0.09525 0.09453 0.09477 0.09430 0.09736 0.09978 0.09828 0.09828 0.09736 0.09758 0.09781 0.09736

0.004 0.043 0.660 0.063 0.019 0.275 0.615 0.019 0.144 0.943 0.615 0.192 0.075 0.579 0.392 0.011 0.228 0.920 0.014 0.396 0.014 0.114 0.975 0.201 0.014 0.018 0.950 0.100

426 8

Membership

4

8

Table VIII. LSD multiple comparison (the effects of combinations of VPEs on SOVC)

Note: *p , 0.05

Figure 4. Main effect of VPE combinations on membership

p ¼ 0.048 , 0.05). According to the LSD analytical results, the immersion effect of group 8 (Escapist þ Social) was significantly better than the effect of group 1 (Education þ Information), group 2 (Education þ Social) and group 6 (Aesthetics þ Social) ð p ¼ 0:0090:0390:024 , 0:05Þ; the membership effect of group 8 (Escapist þ Social) was significantly better than only the effect of group 4 (Entertainment þ Social)

ð p ¼ 0:024 , 0:05Þ: Besides, in sequence 2 (machine ! interpersonal), there was no statistically significant difference for the eight combinations of VPEs relative to SOVC ðWilks’sl ¼ 0:953; F ¼ 1:294; p ¼ 0:205 . 0:05Þ: Therefore, the results partially support H2a but do not support H2b (Table IX). The interaction effect between the degree of PSI and experience combinations was tested by two-way MANOVA. The sample was divided into high and low groups on the basis of a median split on PSI ðmedian ¼ 3:80Þ; this approach follows past research (Puttevu, 2008; Martin et al., 2005). As shown in Table X, results of the MANOVA indicate a significant main effect for PSI (Wilks’s l ¼ 0.938, F ¼ 25.385, p ¼ 0.000 , 0.05) and the eight VPEs combinations (Wilks’s l ¼ 0.956, F ¼ 2.503, p ¼ 0.002 , 0.05). More important, a significant interaction effect was observed between PSI and the eight VPEs combinations ðWilks’sl ¼ 0:951; F ¼ 2:812; p ¼ 0:000 , 0:05Þ

Effects of virtual experiences

427

Figure 5. Main effect of VPE combinations on immersion

Figure 6. Interaction effect between VPE combination and VPE sequence on membership

Figure 7. Interaction effect between VPE combination and VPE sequence on immersion

INTR 21,4

(Figure 8 and Figure 9). Because the interaction effect of PSI by experience combinations was significant, the means of the cells were examined by performing a post hoc LSD test (Table XI). For the high degree of PSI, the LSD multiple comparison results show that group 8 (Escapism þ Social) generated the strongest positive effect on SOVC. For the low degree of PSI, the immersion effect of group 5 (Aesthetics þ

428 Sequence Table IX. Results of MANOVA for VPEs sequence and VPEs combinations

Table X. Results of MANOVA for PSI and VPEs combinations

Figure 8. Interaction effect between VPE combination and PSI on membership

Figure 9. Interaction effect between VPE combination and PSI on immersion

Interpersonal ! Machine Machine ! Interpersonal

VPEs combinations VPEs combinations

Wilks’s l

F-value

p-value

0.942 0.953

1.715 1.294

0.048 * 0.205

Note: *p , 0.05

PSI VPEs combinations PSI * VPEs combinations Note: *p , 0.05

Wilks’s l

F-value

p-value

0.938 0.956 0.951

25.385 2.503 2.812

0.000 * 0.002 * 0.000 *

SOVC High PSI

Low PSI

Immersion

Group I

Group J

Mean (I – J)

Standard error

p-value

8

1 2 3 4 5 6 7 1 2 3 4 5 6 7 1 2 3 4 6 7 8 1 2 3 4 6 7 8

0.7510 * 0.6352 * 0.3256 * 0.3769 * 0.4698 * 0.5734 * 0.3815 * 0.5844 * 0.2823 * 0.3378 * 0.4588 * 0.4773 * 0.3027 * 0.3801 * 0.2730 * 0.2747 * 0.0666 0.0416 0.2796 0.0400 0.0028 0.0380 0.1341 0.1534 0.3127 * 0.0127 0.0678 0.1338

0.15984 0.15908 0.15764 0.15630 0.15764 0.15696 0.15630 0.14064 0.13997 0.13871 0.13753 0.13871 0.13811 0.13753 0.13731 0.13353 0.13496 0.13353 0.13423 0.13353 0.13496 0.15073 0.14658 0.14815 0.14658 0.14735 0.14658 0.14815

0.000 0.000 0.040 0.016 0.003 0.000 0.015 0.000 0.044 0.015 0.001 0.001 0.029 0.006 0.048 0.040 0.622 0.755 0.038 0.765 0.984 0.801 0.361 0.301 0.034 0.931 0.644 0.367

Membership

8

Immersion

5

Membership

5

Note: *p , 0.05

Information) was significantly better than only the effect of group 4 (Entertainment þ Social) ð p ¼ 0:034 , 0:05Þ; also, the membership effect of group 5 (Aesthetics þ Information) was significantly better than the effect of group 1 (Education þ Information), group 2 (Education þ Social) and group 6 (Aesthetics þ Social) (p ¼ 0.048 0.040 0.038 , 0.05). Consequently, H3a was confirmed, H3b was partial accepted. Discussion This study verifies that VPEs affects SOVC. Experimental results show that different VPEs combinations (interpersonal and machine) produce different SOVC. Koh and Kim (2003) note that SOVC is affected by the entertainment offered by a community and a direct relationship exists between entertainment and virtual experiences. We infer that different sequences of virtual experience have different effects on SOVC. It is true that there was no interaction effect among the VPEs sequences, the VPEs combinations and SOVC; however, according to main effect of VPEs sequences by MANOVA, the eight VPEs combinations in sequence 1 (interpersonal ! machine) resulted in different SOVC, and group 8 (Escapism þ Social) had a greater effect on

Effects of virtual experiences

429

Table XI. LSD multiple comparison

INTR 21,4

430

SOVC than other combinations. But in sequence 2 (machine ! interpersonal), the main effect was not significant. Maybe in sequence 2 (machine ! interpersonal), the WOM (interpersonal VPEs) resulting from community interaction generates a greater effect than machine-based virtual experiences, decreasing the negative effects of advertisement and, hence, causing the differences (1983) of SOVC are not significant. This result is consistent with the results in Traylor and Mathias’s research, when advertisements were consumers’ only source of information, the effect is generally negative. When advertisements were coupled with WOM as a source of information, much of the negative effect is attenuated. According to Koh and Kim (2003), web site entertainment affects SOVC that follows from either machine interaction or interpersonal interaction. The forms of web sites and communities conform exactly to interactivity type. Thus, consumers who are engaged in different levels of PSI have varying interaction demands for different web sites and communities. Hence, different virtual experience combinations of interactivity and vividness exert different effects on SOVC. Experimental results indicate that interaction effects exist among the degrees of PSI, the combination of VPEs and SOVC. In a further analysis, group 8 (escapism þ social) has the best effect on membership and immersion under a high degree of PSI. However, under a low degree of PSI, group 5 (aesthetics þ information) has the best effect partially. Theoretical implications The findings of our study have some important theoretical implications for relevant research. First, this study based on the category of experiences by Pine and Gilmore (1999), proposing the four forms of machine VPEs combined with the telepresence concept (Steuer, 1992) and ELM (Petty et al., 1983), employing the model and verifying the effect. Second, according to the current study’s classification of machine VPEs and interpersonal VPEs, analyzing the differences of each of the eight VPEs combinations on SOVC. Third, the current study also uses the high or low degrees of PSI as a moderating variable, confirming the moderate effect between different combinations of VPEs and SOVC. Managerial implications Based on the experiment results, certain managerial implications can be drawn from this study. First, the results can be applied to the customer management on VPEs; manufacturers can refer to such combinations, plan suitable marketing strategies and thereby reinforce the SOVC for consumers in the process of consumer’s purchase decision. For example, in the process of pre-purchase, the demands of consumers can be irritated by the appropriate VPEs combinations; in the process of post-purchase, reinforcing the SOVC for consumers can reduce post-purchase dissonance and enhance the loyalty. Second, the characteristics of PSI provide manufacturers with various marketing strategies for customization of VPEs. The manufacturers of digital camera can regard PSI as the market segmentation variable in order to enhance the brand SOVC and loyalty. Consumers with high degrees of PSI are presented as high interaction in web sites and communities so that marketers can strengthen consumers’ sense of virtual community effects. Conversely, consumers with low degrees of PSI should be presented with web sites featuring high vividness and low interaction. So that such consumers’ sense of community improves, the combination of aesthetics and

information can function to enhance member attention and sense of membership. Third, consumers with a high degree of PSI tend to trust and recommend information; such consumers are assumed to have a positive influence on virtual communities. Hence, when there are new digital cameras going on the market, marketers can design the VPEs combination of escapism and social on their web site, providing the new product to try out or reward aimed at the members with high PSI. Because they have greater SOVC, these members can help the company to carry on the marketing as opinion leaderships, producing better persuasion effect on potential consumers. Limitations and suggestions In this study, the machine VPEs (the four types of web sites) and interpersonal VPEs (the two types of communities) were designed just for this experiment. Our evaluation of SOVC may be ineffective; thus, we suggest that future research cooperate with existing communities or by using different store atmosphere in web retailing (Manganari et al., 2009), the SOVC may be improved to increase external validity. This study examines the effect of VPEs only; future research can consider the different experience combination among VPEs, IDEs and Des. As an online questionnaire is adopted in this study, future research can use different methods and compare experimental results obtained with different research methods. This study focuses on one specific product. Future studies can focus on different products and conduct different comparisons, such as the differences among search product, experiential product and credence product. This study does not address demographic variables and lifestyles, or emotional state and social factors suggested by Dennis et al. (2009). Thus, future studies can further analyze the background and characteristics of each consumer. References Armstrong, A. and Hagel, J. III (1996), “The real value of on-line communities”, Harvard Business Review, Vol. 74 No. 3, pp. 134-41. Ballantine, P.W. and Martin, B.A.S. (2005), “Forming parasocial relationships in online communities”, Advances in Consumer Research, Vol. 32 No. 1, pp. 197-202. Barnes, S. and Mattsson, J. (2008), “Brand value in virtual worlds: an axiological approach”, Journal of Electronic Commerce Research, Vol. 9 No. 3, pp. 195-206. Barnes, S. and Pressey, A.D. (2011), “Who needs cyberspace? Examining drivers of needs in second life”, Internet Research, Vol. 21 No. 3. Bei, L.T., Chen, E.Y.I. and Widdows, R. (2004), “Consumers’ online information search behavior and the phenomenon of search vs experience products”, Journal of Family and Economic Issues, Vol. 25 No. 4, pp. 449-67. Blanchard, A.L. and Markus, M.L. (2004), “The experienced sense of a virtual community: characteristics and processes”, ACM SIGMIS Database, Vol. 35 No. 1, pp. 64-79. Burke, R.R. (1997), “Do you see what I see? The future of virtual shopping”, Journal of the Academy of Marketing Science, Vol. 25 No. 4, pp. 352-60. Chiou, W.B., Wan, C.S. and Lee, H.Y. (2008), “Virtual experience vs brochures in the advertisement of scenic spots: how cognitive preferences and order effects influence advertising effects on consumers”, Tourism Management, Vol. 29 No. 1, pp. 146-50.

Effects of virtual experiences

431

INTR 21,4

432

Cho, C.H. (1999), “How advertising works on the WWW: modified elaboration likelihood model”, Journal of Current Issues and Research in Advertising, Vol. 21 No. 1, pp. 33-50. Daugherty, T., Li, H. and Biocca, F. (2008), “Consumer learning and the effects of virtual experience relative to indirect and direct product experience”, Psychology and Marketing, Vol. 25 No. 7, pp. 568-86. Dennis, C., Merrilees, B., Jayawardhena, C. and Wright, L.T. (2009), “E-consumer behavior”, European Journal of Marketing, Vol. 43 Nos 9/10, pp. 1121-39. Ducoffe, R.H. (1996), “Advertising value and advertising on the web”, Journal of Advertising Research, Vol. 36 No. 5, pp. 21-35. Fornell, C. and Larcker, D.F. (1981), “Evaluating structural equation models with unobservable and measurement error”, Journal of Marketing Research, Vol. 18 No. 1, pp. 39-50. Fortin, D.R. and Dholakia, R.R. (2005), “Interactivity and vividness effects on social presence and involvement with a web-based advertisement”, Journal of Business Research, Vol. 58 No. 3, pp. 387-96. Franke, G.R., Huhmann, B.A. and Mothersbaugh, D. (2004), “Information content and consumer readership of print ads: a comparison of search and experience products”, Journal of the Academy of Marketing Science, Vol. 32, pp. 20-31. Girard, T. and Dion, P. (2010), “Validating the search, experience and credence product classification framework”, Journal of Business and Research, Vol. 63, pp. 1079-87. Girard, T., Korgaonkar, P. and Silverblatt, R. (2003), “Relationship of type of product, shopping orientations and demographics with preference of shopping on the Internet”, Journal of Business Psychology, Vol. 18 No. 1, pp. 101-20. Griffith, D.A. and Chen, Q. (2004), “The influence of virtual direct experience (VDE) on on-line ad message effectiveness”, Journal of Advertising, Vol. 33 No. 1, pp. 55-68. Hoch, S.J. and Deighton, J. (1989), “Managing what consumers learn from experience”, Journal of Marketing, Vol. 53 No. 2, pp. 1-20. Hoerner, J. (1999), “Scaling the web: a parasocial interaction scale for world wide web sites”, in Schumann, D.W. and Thorson, E. (Eds), Advertising and the World Wide Web, Lawrence Erlbaum, Mahwah, NJ, pp. 135-47. Hoffman, D.L. and Novak, T.P. (1996), “Marketing in hypermedia computer-mediated environments: conceptual foundations”, Journal of Marketing, Vol. 60 No. 3, pp. 50-68. Horton, D. and Wohl, R.R. (1956), “Mass communication and parasocial interaction: observations on intimacy at a distance”, Psychiatry, Vol. 19, pp. 215-9. Huang, P., Nicholas, H.L. and Mitra, S. (2009), “Searching for experience on the web: an empirical examination of consumer behavior for search and experience goods”, Journal of Marketing, Vol. 73, pp. 55-69. Keng, C.-J. and Lin, H.-Y. (2006), “Impact of telepresence levels on Internet advertising effects”, CyberPsychology and Behavior, Vol. 9 No. 1, pp. 82-94. Keng, C.-J. and Ting, H.-Y. (2009), “The acceptance of blogs: using a customer experiential value perspective”, Internet Research, Vol. 19 No. 5, pp. 479-95. Klein, L.R. (1998), “Evaluating the potential of interactive media through a new lens: search versus experience goods”, Journal of Business Research, Vol. 41 No. 3, pp. 195-203. Koh, J. and Kim, Y.G. (2003), “Sense of virtual community: a conceptual framework and empirical validation”, International Journal of Electronic Commerce, Vol. 8 No. 2, pp. 75-93. Li, H., Daugherty, T. and Biocca, F. (2001), “Characteristics of virtual experience in electronic commerce: a protocol analysis”, Journal of Interactive Marketing, Vol. 15 No. 3, pp. 13-30.

Li, H., Daugherty, T. and Biocca, F. (2002), “Impact of 3-D advertising on product knowledge, brand attitude, and purchase intention: the mediating role of presence”, Journal of Advertising, Vol. 31 No. 3, pp. 43-58. Lin, H. (2007), “The role of online and offline features in sustaining virtual communities: an empirical study”, Internet Research, Vol. 17 No. 2, pp. 119-38. McMillan, D.W. and Chavis, D.M. (1986), “Sense of community: a definition and theory”, Journal of Community Psychology, Vol. 14 No. 1, pp. 6-23. Mandel, N. and Johnson, E.J. (2002), “When web pages influence choices: effects of visual primes on experts and novices”, Journal of Consumer Research, Vol. 29 No. 2, pp. 235-45. Manganari, E.E., Siomkos, G.J. and Vrechopoulos, A.P. (2009), “Store atmosphere in web retailing”, European Journal of Marketing, Vol. 43 Nos 9/10, pp. 1140-53. Marks, L.J. and Kamins, M.A. (1988), “The use of product sampling and advertising: effects of sequence of exposure and degree of advertising claim exaggeration on consumers’ belief strength, belief confidence and attitudes”, Journal of Marketing Research, Vol. 25 No. 3, pp. 266-81. Martin, B.A.S., Sherrard, M.J. and Wentzel, D. (2005), “The role of sensation seeking and need for cognition on web-site evaluations: a resource matching perspective”, Psychology & Marketing, Vol. 22 No. 2, pp. 109-26. Mathwick, C., Malhotra, N. and Edward, R. (2001), “Experiential value: conceptualization, measurement, and application in the catalog and Internet shopping environment”, Journal of Retailing, Vol. 77 No. 1, pp. 39-56. Millar, M.G. and Millar, K.U. (1996), “The effect of direct and indirect experience on affective and cognitive responses and the attitude-behavior relation”, Journal of Experimental Social Psychology, Vol. 32 No. 6, pp. 561-79. Nelson, P. (1970), “Information and consumer behavior”, Journal of Political Economy, Vol. 78 No. 2, pp. 311-29. Nelson, P. (1974), “Advertising as information”, Journal of Political Economy, Vol. 82 No. 4, pp. 729-54. Nicholson, C.Y., Compeau, L.D. and Sethi, R. (2001), “The role of interpersonal liking in building trust in long-term channel relationships”, Academy of Marketing Science, Vol. 29, Winter, pp. 3-15. Novak, T.P., Hoffman, D.L. and Yung, Y.F. (2000), “Measuring the customer experience in online environments: a structural modeling approach”, Marketing Science, Vol. 19 No. 1, pp. 22-42. Ozok, A.A. and Komlodi, A. (2009), “Better in 3-D? An empirical investigation of user satisfaction and preferences concerning two-dimensional and three-dimensional product representations in business-to-consumer e-commerce”, International Journal of Human-Computer Interaction, Vol. 25 No. 4, pp. 243-81. Ozuem, W., Howell, K.E. and Lancaster, G. (2008), “Communicating in the new interactive marketplace”, European Journal of Marketing, Vol. 42 Nos 9/10, pp. 1059-83. Petty, R.E., Cacioppo, J.T. and Schumann, D. (1983), “Central and peripheral routes to advertising effectiveness: the moderating role of involvement”, Journal of Consumer Research, Vol. 10 No. 2, pp. 135-46. Pine, B.J. and Gilmore, J.H. (1999), The Experience Economy: Work is Theater and Every Business a Stage, Harvard Business Press, Boston, MA. Puttevu, S. (2008), “Consumer responses toward sexual and nonsexual appeals”, Journal of Advertising, Vol. 37 No. 2, pp. 57-69.

Effects of virtual experiences

433

INTR 21,4

434

Rubin, A.M. and Step, M.M. (2000), “Impact of motivation, attraction and parasocial interaction on talk radio listening”, Journal of Broadcasting and Electronic Media, Vol. 44 No. 4, pp. 635-54. Rubin, A.M., Perse, E.M. and Powell, R.A. (1985), “Loneliness, parasocial interaction and local television news viewing”, Human Communication Research, Vol. 12 No. 2, pp. 155-80. Srinivasan, S.S., Anderson, R. and Ponnavolu, K. (2002), “Customer loyalty in e-commerce: an exploration of its antecedents and consequences”, Journal of Retailing, Vol. 78 No. 1, pp. 41-50. Stephens, D.L., Hill, R.P. and Bergman, K. (1996), “Enhancing the consumer-product relationship: lessons from the QVC home shopping channel”, Journal of Business Research, Vol. 37 No. 3, pp. 193-200. Steuer, J. (1992), “Defining virtual reality: dimensions of determining telepresence”, Journal of Communication, Vol. 42 No. 4, pp. 73-93. Suh, K.S. and Lee, Y.E. (2005), “The effects of virtual reality on consumer learning: an empirical investigation”, MIS Quarterly, Vol. 29 No. 4, pp. 673-97. Thorbjornsen, H., Supphellen, M., Nysveen, H. and Pedersen, P.E. (2002), “Building brand relationships online: a comparison of two interactive applications”, Journal of Interactive Marketing, Vol. 16 No. 3, pp. 17-34. Voorveld, H.A.M., Neijens, P.C. and Smit, E.G. (2009), “Consumers’ responses to brand web sites: an interdisciplinary review”, Internet Research, Vol. 19 No. 5, pp. 535-65. Wooldridge, B.R. (1999), “The impact of communal behaviors on the judgment of service quality”, Doctoral dissertation, Louisiana State University, Shreveport, LA. Wu, G.M. (2002), “The role of perceived interactivity in interactive ad processing”, Doctoral dissertation, the University of Texas at Austin, Austin, TX. Zhou, T. (2011), “Understanding online community user participation: a social influence perspective”, Internet Research, Vol. 21 No. 1, pp. 67-81. Further reading Traylor, M. and Mathias, A. (1983), “The impact of TV advertising versus word of mouth on the image of lawyers: a projective experiment”, Journal of Advertising, Vol. 12 No. 4, pp. 42-9. Corresponding authors Ching-Jui Keng can be contacted at: [email protected] and [email protected] Hui-Ying Ting can be contacted at: [email protected] Ya-Ting Chen can be contacted at: [email protected]

To purchase reprints of this article please e-mail: [email protected] Or visit our web site for further details: www.emeraldinsight.com/reprints

Copyright of Internet Research is the property of Emerald Group Publishing Limited and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use.