TY - JOUR
AU - Kim, Seongcheol
AB - Abstract 1Multihop communications will be a key element and will play an important role in 4G wireless networks. To deliver multihop functionality successfully, user participation is imperative. Unfortunately, users of a network may not wish to forward data for others and consume their own scarce resources without rewards. There is an inherent tension between an individual's rationality and whole network welfare, reducing the viability of multihop communications. To understand better why users may be resistant to multihop communications and the sources of such resistance, we propose 4 types of perceived risk that generate such resistance. Based on an online survey, this study found new theoretical links among perceived risk, resistance and participation intention. Introduction 4G (Fourth Generation) mobile communication systems are under study and trial in various fields, with researchers working to establish a global standard, which is expected to be introduced around 2010. 4G evolution is expected to provide ubiquitous environments that enable users to accomplish various tasks, access information, or communicate with others anywhere, anytime, and from any device. Thus 4G is intended to create a new heterogeneous and integrated network based on an open system approach (Chlamtac et al., 2003). Thus, a new architectural approach may be required to attain the full expected capability of 4G mobile communications; for instance, a noninfrastructure-based network could be a good solution. This noninfrastructure-based network inevitably requires multihop transmission that relays packets through other users’ devices. It is rational to think that multihop functionality will be a key element and play an important role in 4G networks. To deliver multihop functionality successfully, user participation is imperative. Unfortunately, users in the network may not wish to forward data for others and consume their own scarce resources without rewards. There is an inherent tension between an individual's rationality and whole network welfare, reducing the viability of multihop communications. The purpose of this study is to examine why users may be resistant to multihop communications and what the sources of this resistance are. By considering both multihop characteristics and user rationality, we propose four types of perceived risk—expected network quality concerns, expected privacy concerns, expected lack of cohesion, and expected resource sacrifice—that provoke resistance to participation in multihop communications. The rest of the study is organized as follows. Section 2 describes the characteristics of multihop in 4G wireless networks. Section 3 reviews theories of users’ technology acceptance and introduces a research model. Section 4 presents the research design, and Section 5 shows data analysis and results. Finally, the discussion and conclusion are provided in Section 6. Multihop Communications in 4G Networks Advent of 4G Networks The mainstream mobile communication technologies have evolved from precellular technology (0G) to 3G technology. With the introduction of analog cellular services, 1G got started and users could place their own calls and sustain conversations seamlessly while moving. 2G is digital cellular technology that allows for more communication channels in the same spectrum. The third-generation (3G) mobile technology, which is characterized as having such features as more bandwidth, a higher data rate and global standards, has been adopted in many countries because true broadband data speeds are required in wireless networks (Shelper, 2005). Traditional ideas of public network infrastructure and licensed spectrum have been applied to 1G, 2G, and 3G networks. On the other hand, 4G technology, which is beyond traditional 1G, 2G, and 3G mobile technologies, does not follow the history of these traditional technologies nor their infrastructure-based licensing and service models. In order to create ubiquitous communication environments, utilization of multiple alternative technologies is expected to be a key factor in the realization of 4G networks (Suliman et al., 2005). Consequently, 4G networks will contain noninfrastructure-based networks, which require multihop transmission among users’ mobile devices. 4G Multihop Applications It may be attractive to utilize a 4G network in areas where there is limited or nonexistent coverage of traditional network infrastructure. Thus 4G applications are sometimes generated and terminated locally when users are nearby. For example, to improve battlefield communications and survivability, tactical network applications have utilized multihop characteristics (Chlamtac et al., 2003). The U.S. Army employed multihop communications in the 2003 Iraq war to establish wireless networks and exchange information among combat vehicles. The dynamic situation on battlefields makes it hard for the military to rely on access to a fixed preplaced network infrastructure. In addition, pure mobile communications have the limitation that radio signals are subject to interference and that radio frequencies higher than 100 MHz rarely propagate beyond line of sight (Freebersyser & Leiner, 2001). Since 4G networks using multihop communication have several advantages, other nonmilitary applications of 4G have attracted considerable attention. For example, multihop applications rather than a fixed infrastructure can be deployed in cases of earthquake or hurricane. In addition, multiusers can play network games without traditional network connectivity. One pertinent 4G application could be in vehicles. Every vehicle on the road is controlled by individual drivers with different interests. Without a central authority, each driver could generate and exchange information. Drivers who forget to bring music CDs will not need to forgo music. With multihop communications, they will be able to search for and download their favorite music files from neighboring car systems. Furthermore, car-to-car communications would enable drivers to exchange information regarding the location of gas stations, restaurants, or public hotspots. On the other hand, certain types of information may only be meaningful by aggregating information from multiple neighboring cars. For example, a driver could guess the road traffic condition by aggregating information from vehicles in a number of locations. Advantages and Obstacles to Multihop Communications 4G networks contain imperative changes that are mainly incurred by introducing multihop communications to an existing communication mechanism. Therefore, understanding the characteristics of multihop communications is critical for further investigation of 4G network possibilities and limitations. Overall, multihop communications outperform single-hop (traditional) communications. First, there is the feature of adaptability (Ilyas, 2003). Under multihop communications, users would not necessarily rely on fixed infrastructure such as APs (Access Points) or base stations. This characteristic is especially useful under the emergency situation where existing infrastructure is destroyed. Second, network coverage could be increased without adding new infrastructure (Maeshima et al., 2003). It is possible to serve those who are located outside of the service area with the help of neighboring users’ devices. Third, there is the advantage of spatial reuse (Salem et al., 2004). Data transmitted by neighboring devices requires small-radius transmissions. Reuse of the same frequency in a different coverage area could enhance spectral efficiency and increase the available transmission rate (Li et al., 2002). Fourth, it is possible to reduce the required transmission power in a multihop cell because several low-power links are established among neighboring users’ devices (Li et al., 2002). As a result of low-power links, total interference could be also reduced. The energy savings afforded by multihop data transmitting would help to conserve battery resources of mobile devices. Despite the merits discussed above, prospective multihop applications must cope with several practical problems. Technical issues Because multihop communications do not have central authorities, heavy information exchange in one part could impair the optimization of the whole network. As multihop networks can be formed and disbanded according to necessity, power allocation and spectrum allocation should be controlled and changed accordingly. Furthermore, security vulnerabilities are accentuated in multihop communication because any unauthorized mobile user's device can participate in the network and potentially serve as a traffic router (Chlamtac et al., 2003). Data delivery in free space exposes the multihop network to attack and eavesdropping. These technical challenges may degrade service quality and eventually deter users from participating in multihop communications. Privacy issues Privacy concerns have become significant because of the free participation and anonymity of users. Some information may be relevant to users, such as their location. However, repeated exchange with an unknown neighboring user can allow matching of data and being identified by others. Consequently, even tiny correlations among the data may reveal useful information to others. Exposure of users’ locations and application information to others may increase privacy concerns such as reuse without consent. Behavioral issues Each mobile device is operated by a self-interested and rational user in multihop communications. Thus, it is natural that human users are not motivated to participate without “common goals” or “rewards” for their contribution. As shown in the vehicular scenario, a user may feel it better to turn off the device and be alone without disturbance to avoid competition for limited resources. Otherwise, a user may maximize a particular function at the expense of another user's device (Conti et al., 2004). Because multihop communications have not only advantages but also obstacles, it is worthwhile examining why users may be resistant to participating in multihop communications. Literature Review Technology Acceptance User acceptance of new technology is one of the richest research areas in the fields of information systems (IS) and new media. Thus studies in these areas have yielded many competing models based on technology, psychology, and sociology (Venkatesh et al., 2003). Theoretical effort was initially derived from Fishbein and Ajzen's (1975) theory of reasoned action (TRA), which indicated that individual behavior is a function of both an individual's attitude toward a specific behavior and the social influences and norms surrounding it. The technology acceptance model (TAM), which evolved from TRA, argued that perceived usefulness and perceived ease of use predict technology usage through the mediating variables of attitude and intention (Davis et al., 1989). According to the diffusion of innovation theory (DIT), adoption is a function of perceived attributes of innovations, social norms, and individual characteristics (Rogers, 1995). These theories are repeatedly applied to examine not only traditional technologies (e.g., personal computers, the Windows operating system, automatic teller machines, and e-mail) but also new or innovative ones such as mobile commerce (Khalifa & Cheng, 2002), wireless Internet (Lu et al., 2003; Kim et al., 2007), and mobile marketing (Bauer et al., 2005). However, these theories have paid scant attention to resistance that would prevent an individual from using a technology. There may be a situation in which an individual wants to use technology but hesitates because of unreliable technological quality or lack of resources. Though some studies have included disconfirmation or computer anxiety as minor belief constructs in models, they dealt with those constructs as black boxes and have been unable to investigate what actually makes users hesitate. In addition, recent studies such as that of Pavlou et al. (2007) investigated the sources of uncertainty in technology acceptance but did not provide a suitable model to explain user participation in a very dynamic context such as multihop communications. Thus there is a need to include proper constructs and their sources in order to provide a better explanation of participation in multihop communications. Resistance Resistance refers to a normal psychological attitude in which the perceived consequences (e.g., loss of power) are risky (Ang & Pavri, 1994). However, a user's resistance to new technology or service has received little attention in the technology acceptance literature. Although researchers acknowledge the importance of resistance, they treat it as a black box. Most researchers have focused on successful change by assuming that every innovation is beneficial and should be accepted by all members of a social system (Ram, 1987). This is mainly caused by the biased idea that all innovations improve the prevailing situation for the majority of users (Szmigin & Foxall, 1998). However, resistance should be understood as a natural factor of a decision process rather than as a necessary factor leading to nonadoption (Kuisma et al., 2007). It is frequently stimulated when users have a tendency to retain the existing practice or perceive risk related to new events. Thus, resistance could be applied in the multihop context. Multihop networks are an unfamiliar communication method in that users will be not only beneficiaries (source or destination users) but also routers who relay information to others. Participation by unknown users rather than existing telecommunication operators may trigger resistance by all users to multihop networks. Therefore, resistance as an attitude should be included in a model that explains users’ participation in multihop communications. Empirical tests of technology acceptance models suggest that intention is a strong and significant predictor of actual behavior. As multihop communications are not yet widely commercialized, we use intention as an estimate of actual participation. Regarding the relation between attitudes and intention, most models indicate a positive link. However, it is expected that there is a negative relation between resistance (attitude) and participation intention. For example, previous studies have shown that uncertainty, a kind of negative attitude, decreased participation intention in online transactions (Pavlou et al., 2007). Based on the above discussion, we hypothesize the relation between resistance and participation intention as follows. H1: Resistance will have a negative effect on participation intention in multihop communications. Perceived Risk Changes that threaten core values or existing habits are important causes of resistance (Arnould et al., 2004). Sheth (1981) argued that perceptions associated with physical, social, economic, or functional risk are the most useful constructs for resistance. In a similar context, Ram (1987) categorized three sources of resistance: innovation characteristics, consumer characteristics, and characteristics of the propagation mechanism. However, the propagation mechanism is deleted from resistance sources because it is considered a barrier factor from the producer perspective (Ram, 1987). Thus, in order to capture resistance to multihop communications, it is reasonable to consider risk in terms of both technology and user characteristics. The risk concept has frequently appeared in research on customer behavior (Bauer, 1967; Peter & Ryan, 1976; Mitchell, 1992; Dowling & Staelin, 1994). It has also been adopted in research on customer behavior related to information technology (Germunden, 1985; Hoffman et al., 1999; Featherman, 2001; Pavlou, 2001). Although the risk concept in technology acceptance research has frequently appeared, it has not received attention as a major construct. Perceived risk is defined as the extent to which a person feels uncertainty regarding the possible negative consequences of using a technology (Featherman & Pavlou, 2003). Resistance to multihop communications can be induced by perceived risk because resistance as a negative attitude is stimulated when users tend to perceive risk related to new technology. Accordingly, we assume that perceived risk is a critical source of resistance to multihop communications. Furthermore, considering the various issues in multihop communications, we specify perceived risk by proposing four types of perceived risk: (1) expected network quality concerns, (2) expected privacy concerns, (3) expected lack of cohesion, and (4) expected resource sacrifice. Expected network quality concerns (ENQCs) As a generic performance measure, quality has numerous definitions. According to ITU-T, QoS (Quality of Service) is defined as the collective effort of service performance, which determines the degree of satisfaction of users of a service. The European Union's R&D in Advanced Communications technologies in Europe (RACE) defines QoS as a set of user-perceivable attributes of that which makes a service what it is (RACE, 1994). Even though QoS is a multidimensional concept, users cannot separate it into subfactors, and they consider it a single factor (Koivisto & Urbaczewski, 2004). Because of free entry and exit in multihop communications, the most easily perceivable quality could be network quality. Throughout previous studies, reliability, efficiency, predictability, and satisfaction are the four main factors measuring user-perceived network quality (Saliba et al., 2005). Thus, in the context of multihop networks, expected network quality concerns could be described as a user's subjective belief about the reliability, efficiency, predictability, and satisfaction of networks. In the case of traditional communications, there is a dedicated infrastructure such as a base station for controlling network quality. However, multihop communications would not always depend on a central infrastructure for network quality management. As a result, a destination user eventually receives poor outcomes such as service delays or network shutdown. If users continuously experience or expect poor network quality, they may no longer participate in multihop communications. Thus, unsecured network quality would create a resistant attitude toward multihop because users would be unable to trust network services fully. Therefore, this study proposes the second hypothesis as follows. H2: Expected network quality concerns will have a positive effect on resistance to multihop communications. Expected privacy concerns (EPCs) Privacy issues in transactions are raised by participating users rather than by the delivered products or services (Pavlou et al., 2007). For example, consumers are worried about seller opportunism such as imposture or contract default because a seller's behavior cannot be easily monitored or guaranteed (Pavlou et al., 2007). Expected privacy concerns are defined as a user's belief about other participants’ uncertainty in protecting personal data from improper use, disclosure to third parties, and secondary use without consent. Under multihop communications, it is inevitable that neighboring users’ information will be exchanged to transmit requested data. When services are delivered, a user may face a dilemma. On one hand, a user must forward data to others, so that they can forward the data toward the destination. On the other hand, this user may not trust any of the other parties and may hesitate to forward the data to preserve privacy (Hong et al., 2006). Because the user has no way of discovering which other users are malicious, he/she might be unsure regarding transactions. In contrast, there is a central authority that senders and receivers can trust in traditional networks. Any participating user, rather than one central authority, can accumulate tiny data over a long period time and can eventually find correlations between them. Mason (1986) mentioned that privacy concerns are stimulated as interaction among participants is increased. Successful multihop communications require dynamic interaction among users. Therefore, innocent-looking data from various sources and repetition can reveal useful information to others. For example, overzealous telemarketers watch every move a target consumer makes and manufacture location information for secondary use without his/her knowledge. Accordingly, we assume that users with expected privacy concerns may have resistant attitudes. H3: Expected privacy concerns will have a positive effect on resistance in multihop communications. Expected lack of cohesion According to Hogg (1992), group cohesion can be explained as a sense of a member's attraction to the group. Cohesion can be strengthened when members of a group share common goals and expectations for results of group performance. Group cohesion produces favorable results such as active commitment in group tasks and improved group performance (Hsu & Lu, 2007). In studies of online game communities, perceived cohesion seems to be a significant determinant of a user's preference for participating in a virtual community (Emirbayerand & Goodwin, 1994; Hsu & Lu, 2007). Traditional networks need only those who send and receive the services to communicate. While a sender and a receiver communicate, social interaction between them builds social intimacy, eventually motivating participation. However, multihop networks introduce intermediary users who may be not directly related to services or service target users. In many instances, group cohesion is thought to be formed from extensive interpersonal interaction (Hsu & Lu, 2007). Lack of social interaction between multihop participants may discourage participation in such networks. In addition, they must compete for limited resources, as well as cooperate. Such inner dynamics of multihop participants all affect cohesion perception (Emirbayerand & Goodwin, 1994). A lack of cohesiveness is likely to form a higher level of resistance (Mukherji et al., 2007). Based on discussion, expected lack of cohesion among users can be considered to enforce a resistant attitude in multihop communications. H4: Expected lack of cohesion will have a positive effect on resistance in multihop communications. Expected resource sacrifice In accordance with the prospect theory (Kahneman & Tversky, 1979), value is assigned to gains and losses with different probability functions. Because the value function is concave for gains and convex for losses, it is normally steeper for losses than for gains even with the same amount. This could be interpreted to indicate that people place more weight on losses than on gains. Similarly, users are frequently more deterred by costs than they are attracted by benefits. It is natural that self-centered users deny others’ requests or pretend their mobile devices to be dead in multihop networks. These behavioral issues are not inherent in any multihop applications for which devices are stationary and attached to large energy and computing sources. Unfortunately, many promising and exciting multihop applications are classified into the resource-constrained category (Goldsmith & Wicker, 2002). People tend to feel and behave negatively when they perceive inevitable damage such as lack of control over their resources. Continuous drains on battery power resulting from intermediating roles in multihop networks may increase psychological risks. For example, a user may be irritated that the lower battery level makes a device unavailable when necessary. As a result, a user may feel that it is rational to turn off the mobile device rather than to allow others to consume limited resources. Therefore, we assume that users of multihop communications tend to have high resistance if they expect more resource sacrifice. H5: Expected resource sacrifice will have a positive effect on resistance to multihop communications. Control variables In order to exclude counterbalancing effects of the four types of perceived risk on resistance, this study considers three control variables: (1) perceived usefulness (PU), (2) perceived ease of use (PEOU), and (3) past experience (PE). As this study explores user participation in multihop communications, we rename the former two variables as expected usefulness (EU) and expected ease of participation (EEOP). EU is defined as the belief that participation will enhance a user's effectiveness, productivity, and performance. EU can be provoked by free services without mainstream telecommunications operators and diverse applications. Prior studies have shown a significant impact on participation (Bagozzi & Dholakia, 2006; Kim et al., 2007). EEOP refers to the degree to which a user believes that participating in multihop communications is effortless (Hsu & Lu, 2007). Here EEOP is interpreted as free of effort when users relay data on behalf of others and use services. The marketing and consumer behavior literature shows a high correlation between PE and actual participation (Kim et al., 2007; Pavlou et al., 2007). Because multihop communications are achieved by users’ mobile devices participating in a peer-to-peer fashion, lack of similar communications (P2P or wireless Internet) experience may impede users from participating in multihop networks. In summary, even though the success of multihop communications depends greatly on user participation, this cannot be guaranteed because of inherent user resistance stemming from perceived risk. The proposed research model in this study integrates the literature and hypotheses described above (Figure 1). Figure 1 Open in new tabDownload slide Research model Figure 1 Open in new tabDownload slide Research model Method Procedures Survey research is frequently used to gauge people's attitudes and individual psychological issues. For this study, the initial sample was comprised of 2,000 randomly selected Korean online users who were notified by an e-mail invitation message linked to an online survey page (http://pollever.com). The participants were offered cyber mileage points as incentives. The invitees were assured that the results would only be used in aggregate to guarantee their anonymity and confidentiality. The online survey was conducted from April 9 to 12, 2007. Because the actual applications of multihop communications are not yet widely commercialized, this study first explained the concept of multihop and its characteristics. In order to assist further understanding, we provided two possible scenarios with visual images before the survey procedure. One was a flea market scenario in which a location-specific advertisement was promoted, and the other was a vehicular communications scenario in which information was shared among drivers. For example, a flea market scenario was illustrated like follows; passers-by could receive commercial advertisements to their mobile phones from neighboring merchants. Such advertisements would be quite convenient for both merchants and some interested users, while the others may be irritated by unwanted interruption. A final total of 1,021 valid responses (51.1% effective response rate) were analyzed in this study. If a respondent partially completed or chose the same item in more than 90 percent of 7-point scale questions, the case was considered haphazard and invalid. Measurement Measurement items for this study were adapted from existing measures but modified for the multihop context. Twenty-seven items were taken directly or adapted from previous literature. Table A in the Appendix shows the measurement items. The survey questionnaire was developed with a 7-point Likert scale from 1 = Strongly disagree to 7 = Strongly agree. Reliability and validity tests were conducted for measurement model verification. Results are shown in Table 1. The reliability test used Cronbach's α and CR (composite reliability). All constructs passed the reliability test because the Cronbach's α value was higher than .70. CR greater than .70 implies that a construct retains both internal consistency and convergent validity. Table 1 Results of reliability and validity test Constructs . No. of items . Cronbach's α . CR . AVE . ENQC 4 .8397 .893 .678 EPC 4 .9202 .944 .807 ELOC 3 .7547 .860 .672 ERS 3 .8365 .902 .754 RES 4 .8613 .906 .707 EU 3 .8376 .902 .755 EEOP 3 .8535 .911 .774 PI 3 .9095 .943 .847 Constructs . No. of items . Cronbach's α . CR . AVE . ENQC 4 .8397 .893 .678 EPC 4 .9202 .944 .807 ELOC 3 .7547 .860 .672 ERS 3 .8365 .902 .754 RES 4 .8613 .906 .707 EU 3 .8376 .902 .755 EEOP 3 .8535 .911 .774 PI 3 .9095 .943 .847 Open in new tab Table 1 Results of reliability and validity test Constructs . No. of items . Cronbach's α . CR . AVE . ENQC 4 .8397 .893 .678 EPC 4 .9202 .944 .807 ELOC 3 .7547 .860 .672 ERS 3 .8365 .902 .754 RES 4 .8613 .906 .707 EU 3 .8376 .902 .755 EEOP 3 .8535 .911 .774 PI 3 .9095 .943 .847 Constructs . No. of items . Cronbach's α . CR . AVE . ENQC 4 .8397 .893 .678 EPC 4 .9202 .944 .807 ELOC 3 .7547 .860 .672 ERS 3 .8365 .902 .754 RES 4 .8613 .906 .707 EU 3 .8376 .902 .755 EEOP 3 .8535 .911 .774 PI 3 .9095 .943 .847 Open in new tab To test validity, CR and AVE (average variance extracted) were calculated. Factor loadings and AVE were tested using the PLS (partial least squares method). AVE measures the percentage of variance captured by a construct by showing the ratio of the sum of the variance captured by the construct and measurement variance. It is acceptable if an AVE exceeds. 50. First, an individual item's factor loading and each construct's CR and AVE were checked using the convergent validity test (Table B in Appendix). The CRs of all eight constructs in Table B were higher than .70, indicating convergent validity. Regarding factor loading, all items exceeded the cutoff point (.70). The smallest AVE (ELOC), which was .671 in this study, was also feasible because it was greater than .50. Accordingly, strong convergent validity was found. In order to examine discriminant validity, a cross loading matrix was created. The cross loading matrix showed correlation between the construct score and the measurement items, suggesting that measurement items load highly on their theoretically assigned constructs and not on other factors. Each item was strongly associated with its respective latent variable; thus, discriminant validity was identified. The cross loading table is attached in Table C. The square root of AVE can be used to examine discriminant validity as well. Diagonal elements should be larger than the entries in the corresponding rows and columns. The correlations among all constructs were well below .70 thresholds, suggesting that all constructs were distinct (see Table D). Because the current study used data from self-reports of participants via an online survey, common method bias might be detected. Common method variance is defined as the artifactual covariance between two variables attributable to the measurement methods rather than to a relationship between the underlying constructs of interest (Campbell & Fiske, 1959). Lindell and Whitney (2001) used a theoretically unrelated construct (termed a marker variable) to filter out unreliable samples. To economize on survey items, this study did not do this. However, the correlation matrix (Table C) did not indicate any highly correlated factors (the highest correlation is r = .644), whereas evidence of common method bias should have resulted in extremely high correlations (r = .90) (Pavlou et al., 2007). Statistical Analysis Data analysis in this study was performed using the PLS method. Since Davis et al. (1989) popularized Structural Equation Models (SEM) by inventing TAM, 81 papers from the five top journals in information systems (Information Systems Research, MIS Quarterly, Management Science, Decision Sciences, and Journal of Management Information Systems) have invoked SEM. SEM methods such as LISREL or AMOS can be generally divided into two types in terms of how they use data (limited vs. full information methods). LISREL is a frequently used method that utilizes full information for confirmatory research. Thus, a large sample size is required, and standard distribution is assumed. In contrast, PLS introduced by Wold (1985) focuses on maximizing the variance of dependent variables explained by independent one. The PLS approach places minimal restrictions on sample size and residual distribution. In addition, the PLS algorithm allows each indicator to vary in its contribution to the composite score of the latent variable instead of assuming equal weights for all indicators of a scale. Tobia (1999) maintained that PLS is useful in screening out negligible factors affecting the dependent variable. Wold (1985) advised that PLS is not suitable for confirmatory testing but should be used for exploration of plausible causality. This study used skewed data for prediction and the exploration of causality despite large samples. For example, more than half of the respondents had experience of person-to-person (P2P) (66.6%) and wireless Internet services (73.3%). This asymmetry of data may be because only Korean users were included in the sample. On the other hand, this study is interested in each specific path coefficient and variance explained rather than overall model fit. Given that this study is an exploratory study of users’ resistance to participation in multihop communications, we used PLS for measurement validation and to test the proposed research model. Results Sample Characteristics Table 2 shows the profile of sample respondents including major demographic information and the characteristics of P2P and wireless Internet service use. Table 2 Profile of sample respondents Profile category . Frequency . Percentage . Gender (N = 1021) Male 661 64 .7 Female 360 35 .3 Age 10–19 12 1 .2 20–29 262 25 .7 30–39 448 43 .9 40–49 214 21 .0 50 or higher 85 8 .3 Occupation Student 103 10 .1 Professional 83 8 .1 Office worker 596 58 .4 Self-employed 103 10 .1 Housewife 86 8 .4 Other 50 4 .9 Education Middle school 10 1 .0 High school 229 22 .4 College 713 69 .8 Advanced degree 69 6 .8 P2P experience Yes 680 66 .6 No 341 33 .4 P2P usage frequency (N = 680) A few times a month or less 277 40 .7 A few times a week 322 47 .4 Every day 81 11 .9 Wireless service experience Yes 748 73 .3 No 273 26 .7 Wireless service usage frequency (N = 748) A few times a month or less 429 57 .4 A few times a week 260 34 .8 Every day 59 7 .9 Profile category . Frequency . Percentage . Gender (N = 1021) Male 661 64 .7 Female 360 35 .3 Age 10–19 12 1 .2 20–29 262 25 .7 30–39 448 43 .9 40–49 214 21 .0 50 or higher 85 8 .3 Occupation Student 103 10 .1 Professional 83 8 .1 Office worker 596 58 .4 Self-employed 103 10 .1 Housewife 86 8 .4 Other 50 4 .9 Education Middle school 10 1 .0 High school 229 22 .4 College 713 69 .8 Advanced degree 69 6 .8 P2P experience Yes 680 66 .6 No 341 33 .4 P2P usage frequency (N = 680) A few times a month or less 277 40 .7 A few times a week 322 47 .4 Every day 81 11 .9 Wireless service experience Yes 748 73 .3 No 273 26 .7 Wireless service usage frequency (N = 748) A few times a month or less 429 57 .4 A few times a week 260 34 .8 Every day 59 7 .9 Open in new tab Table 2 Profile of sample respondents Profile category . Frequency . Percentage . Gender (N = 1021) Male 661 64 .7 Female 360 35 .3 Age 10–19 12 1 .2 20–29 262 25 .7 30–39 448 43 .9 40–49 214 21 .0 50 or higher 85 8 .3 Occupation Student 103 10 .1 Professional 83 8 .1 Office worker 596 58 .4 Self-employed 103 10 .1 Housewife 86 8 .4 Other 50 4 .9 Education Middle school 10 1 .0 High school 229 22 .4 College 713 69 .8 Advanced degree 69 6 .8 P2P experience Yes 680 66 .6 No 341 33 .4 P2P usage frequency (N = 680) A few times a month or less 277 40 .7 A few times a week 322 47 .4 Every day 81 11 .9 Wireless service experience Yes 748 73 .3 No 273 26 .7 Wireless service usage frequency (N = 748) A few times a month or less 429 57 .4 A few times a week 260 34 .8 Every day 59 7 .9 Profile category . Frequency . Percentage . Gender (N = 1021) Male 661 64 .7 Female 360 35 .3 Age 10–19 12 1 .2 20–29 262 25 .7 30–39 448 43 .9 40–49 214 21 .0 50 or higher 85 8 .3 Occupation Student 103 10 .1 Professional 83 8 .1 Office worker 596 58 .4 Self-employed 103 10 .1 Housewife 86 8 .4 Other 50 4 .9 Education Middle school 10 1 .0 High school 229 22 .4 College 713 69 .8 Advanced degree 69 6 .8 P2P experience Yes 680 66 .6 No 341 33 .4 P2P usage frequency (N = 680) A few times a month or less 277 40 .7 A few times a week 322 47 .4 Every day 81 11 .9 Wireless service experience Yes 748 73 .3 No 273 26 .7 Wireless service usage frequency (N = 748) A few times a month or less 429 57 .4 A few times a week 260 34 .8 Every day 59 7 .9 Open in new tab Of the subjects, 64.7% were male and 35.3% were female. The majority of respondents were in their 30s (43.9%) and were mainly office workers (58.4%); 76.6% had higher education including postgraduate, and 66.6% had experienced P2P services. Regarding wireless Internet services, 73.3% answered that they had used them. This sample of online respondents inevitably has some shortcomings. First of all, users with experience of P2P or wireless service were overrepresented even though Koreans are early adopters of new technology. In terms of gender and education, this sample may not represent the whole population well. However, the data from this sample were used in statistical analyses without weighting because the primary purpose of this study is hypothesis testing rather than descriptive investigation (Zhou, 2008). Test of Research Hypotheses As shown in Figure 2, four of five paths were significant at p level of .001, which indicated that hypotheses H1, H2, H4, and H5 were supported. Figure 2 Open in new tabDownload slide PLS results for research model Figure 2 Open in new tabDownload slide PLS results for research model As hypothesized, resistance had a significant negative effect on participation intention (b = −.249, p < .001), validating this paper's key position that the existence of a resistant attitude is an important impediment to participation in multihop networks. Regarding the four types of perceived risk as the sources of resistance, all hypothesized paths to resistance except expected privacy concerns were significant, explaining 8.2% of variance for resistance. In particular, the path coefficient of expected network quality concerns to resistance was the highest (b = .263, p < .001). The effects of expected lack of cohesion (b = .155, p < .001) and expected resource sacrifice (b = .210, p < .001) were statistically significant. However, the effect of expected privacy concerns on resistance was not significant. Table 3 summarizes the result of hypothesis tests. Table 3 Summary of hypotheses testing Hypothesis . Supported . γ . H1: RES ➔ PI Yes −.249* (6.903) H2: ENQC ➔ RES Yes .263* (6.203) H3: EPC ➔ RES No .033 (.613) H4: ELOC ➔ RES Yes .155* (3.482) H5: ERS ➔ RES Yes .210* (4.354) Hypothesis . Supported . γ . H1: RES ➔ PI Yes −.249* (6.903) H2: ENQC ➔ RES Yes .263* (6.203) H3: EPC ➔ RES No .033 (.613) H4: ELOC ➔ RES Yes .155* (3.482) H5: ERS ➔ RES Yes .210* (4.354) * The number in parenthesis is the t-value. † p < .001. Open in new tab Table 3 Summary of hypotheses testing Hypothesis . Supported . γ . H1: RES ➔ PI Yes −.249* (6.903) H2: ENQC ➔ RES Yes .263* (6.203) H3: EPC ➔ RES No .033 (.613) H4: ELOC ➔ RES Yes .155* (3.482) H5: ERS ➔ RES Yes .210* (4.354) Hypothesis . Supported . γ . H1: RES ➔ PI Yes −.249* (6.903) H2: ENQC ➔ RES Yes .263* (6.203) H3: EPC ➔ RES No .033 (.613) H4: ELOC ➔ RES Yes .155* (3.482) H5: ERS ➔ RES Yes .210* (4.354) * The number in parenthesis is the t-value. † p < .001. Open in new tab Discussion and Conclusion Role of Resistance and Perceived Risk As discussed earlier, previous studies seem to be incomplete and potentially misleading in explaining user participation in a very dynamic context like multihop communications because they have paid little attention to resistance and perceived risk that would affect an individual's use of a new technology. The abundance of papers that essentially repeat the positive message of user acceptance has led to the unwarranted belief that people will voluntarily accept new technology (Benbasat & Barki, 2007). However, this study found that resistance as a negative psychological attitude is a key variable in predicting participation intention in multihop communications. It also found that perceived risk plays an important role in explaining users’ resistant attitudes toward multihop communications. Thus the results of this study may enhance understanding of a fundamental set of constructs such as resistance and perceived risk that have been overlooked in the technology acceptance literature. In other words, if technology acceptance theories include resistance and perceived risk in their models, they may provide a better explanation of participation intention and behavior in multihop communications. The proven theoretical link among perceived utilities, positive attitude and behavioral intention in technology acceptance theories should be integrated or complemented by new theoretical links among perceived risk, resistance, and behavioral intention when the advantages and challenges of new technology are mixed and uncertain. In addition, empirical findings of this study may shed light on what actually makes users hesitate to participate in multihop communications. First of all, resistance to multihop communications was found to decrease participation intention. This finding confirms that resistance plays an important role in understanding human behavior toward change. Second, among four types of perceived risk, expected network quality concerns turned out to be the strongest predictor of user resistance. This was not surprising because most literature on network quality stated that it has a great impact on user acceptance. Thus if users cannot find any improvement in network reliability or efficiency, they may stay with existing communication technologies. Third, expected lack of cohesion was shown to have a positive effect on resistance. Because communication is an important social activity, it is important to create cohesion between participants. If users do not feel any interaction with the communicating party, they will be reluctant to participate in multihop communications. Fourth, expected resource sacrifice was found to be a significant source of resistance. This result is mainly related to the limited resources (e.g., battery or memory) of mobile devices for multihop communications. This is consistent with the prospect theory (Conti et al., 2004) indicating that those who contribute expect compensation. However, expected privacy concerns proved not to be a significant source of resistance in this study. This result is inconsistent with previous studies indicating that privacy concerns create negative perceptions and lead to nonuse or delay. This unexpected finding might be explained by the unique characteristics of Korean users, who are less sensitive to privacy issues. Ahn et al. (2001) compared e-commerce adoption behaviors between Americans and Koreans. They found that the impact of privacy risks on adoption of e-commerce is the most important impediment in the U.S., while Koreans do not consider the risks as significant decision factors. This result indicates that privacy is a more significant issue in an individualist culture society like the U.S. than in a collectivist culture society like Korea (Kim, 2005). Implications It is expected that multihop communications, which are noninfrastructure-based networks among users’ mobile devices, will be crucial to so-called 4G networks, the next generation of wireless networks. Although multihop communications face several obstacles and cannot meet all communication requirements, the deployment of this type of wireless technology will provide many opportunities. In the meantime, to take advantage of the opportunities, it is necessary to encourage user participation by mitigating perceived risk and resistance. Thus the results of this study have business as well as social implications. Wireless network operators should make efforts in their businesses to minimize user resistance to multihop communications by reducing users’ perceived risk. First, participating users should be assured of network quality, not only actual quality but also users’ perceived quality (Saliba et al., 2005). As indicated by Caro & Garcia (2007), quality concerns are more closely associated with a service provider's performance such as high market share or enhanced customer loyalty rather than with the network itself. Therefore network operators need to develop a framework to measure quality from the users’ perspective and to pay more attention to users’ experiences in multihop communications. Second, they should create cohesion among users, because in a cohesive atmosphere, there would be less resistance to multihop communications. For example, network operators may organize a form of virtual community for users to share information and build social intimacy or achieve critical mass by utilizing offline cohesion within the same organization, such as a campus or workplace. Third, users of multihop communications should be rewarded for their resource sacrifice. As advanced functions such as MP3 players, cameras, and mobile TVs are added to mobile devices, resource issues will be more critical to 4G networks. If users resist participating in multihop communications, network operators may need to invest significantly in order to establish more 4G infrastructure such as access points or base stations. Thus network operators should design incentive systems that include both monetary incentives (e.g., mileage) and psychological incentives (e.g., reputation) as shown in the P2P sites. There have been many technological approaches to facilitating multihop communications, and user participation has been naturally assumed because of the technology's widely touted merits. However, considering the empirical results of this study, it can be said that placing too much credibility or confidence in technology may create blind spots in 4G evolution if there is no understanding of the several types of perceived risk in, and resistance toward, multihop communications. Therefore, in order to deploy multihop communications successfully, market requirements rather than technological possibilities should be emphasized more strongly, and new business innovations should be created along with technological evolution. From the business point of view, multihop communications may challenge the mainstream telecommunications industry because multihop networks enable users to form their own communications networks, thus in some cases reducing dependency on access networks provided by traditional telecommunications operators. Therefore, if there is enough participation from users, multihop communications may disrupt existing business arrangements in favor of the users rather than the network suppliers, and may present opportunities for new or small wireless network operators to grow. It is also necessary to consider privacy issues in multihop communications even though expected privacy concerns had no effect on resistance in this study. Because privacy issues may become a more serious social problem in other countries, not only network operators but also policy makers should make efforts to cope with such issues in multihop communications. Limitations As with most research efforts, this paper is not without limitations. First, this study includes several positive constructs from technology acceptance theories as control variables rather than independent variables. In order to obtain a more holistic view, positive and negative constructs may need to be considered together in the research model. We believe that developing and testing an integrated research model of users’ participation in multihop communications is an interesting topic for future research. Second, this study is susceptible to sample bias. Because this study focuses only on the Korean context, it is difficult to generalize the findings and implications to other countries. In addition, our survey was designed to be conducted online. Online respondents are likely to be more active and less resistant toward new technology than others. Future research may seek to obtain responses from offline users as well. Finally, in this study, participation intention was used as an estimate of actual participation behavior. As intention and actual participation may have different phases, actual participation in multihop communications can be measured in future studies. Notes 1 Please address correspondence to the second author. References Ahn , J. , P , Jinsoo., & L , Dongwon. ( 2001 ). Risk-focused e-commerce adoption model—A cross-country study . MISRC WP 01–30. Retrieved from http://www.misrc.umn.edu/workingpapers/fullPapers/2001/0130_060101.pdf. Ang , J. , & Pavri , F. ( 1994 ). A survey and critique of the impacts of information technology . International Journal of Information Management , 14 ( 2 ), 122 – 133 . Google Scholar Crossref Search ADS WorldCat Arnould , E. , Price , L., & Zinkhan , G. ( 2004 ). Consumers (2nd ed.). New York : McGraw-Hill . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Bagozzi , R. P. , & Dholakia , U. M. ( 2006 ). Open source software user communities: A study of participation in Linux user groups . Management Science , 52 ( 7 ), 1099 – 1115 . Google Scholar Crossref Search ADS WorldCat Bauer , R. A. ( 1967 ). Consumer behavior as risk taking . In D. F. Cox (Ed.), Risk taking and information handling in consumer behavior (pp. 23 – 33 ). Boston : Harvard University Press . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Bauer , H. H. , Barnes , S. J., Reichardt , T., & Neumann , M. M. ( 2005 ). Driving consumer acceptance of mobile marketing: A theoretical framework and empirical study . Journal of Electronic Commerce Research , 6 ( 3 ), 181 – 192 . Google Scholar OpenURL Placeholder Text WorldCat Benbasat , I. , & Barki , H. ( 2007 ). Quo vadis, TAM? Journal of the Association for Information Systems , 8 ( 4 ), 211 – 218 . Google Scholar OpenURL Placeholder Text WorldCat Campbell , D. T. , & Fiske , D. W. ( 1959 ). Convergent and discriminant validation by the multitrait–multimethod matrix . Psychological Bulletin , 56 ( 2 ), 88 – 105 . Google Scholar Crossref Search ADS WorldCat Caro , L. M. , & Garcia , J. A. M. ( 2007 ). Measuring perceived service quality in urgent transport service . Journal of Retailing and Consumer Services , 14 , 60 – 72 . Google Scholar Crossref Search ADS WorldCat Chlamtac , I. , Conti , M., & Liu , J. J.-N. ( 2003 ). Mobile ad hoc networking: imperatives and challenges . Ad Hoc Networks , 1 ( 1 ), 13 – 64 . Google Scholar Crossref Search ADS WorldCat Conti , M. , Gregori , E., & Maselli , G. ( 2004 ). Cooperation issues in mobile ad hoc networks . Proceedings of the 24th International Conference on Distributed Computing Systems Workshops (ICDCSW ’04), March 23–24, 2004, Tokyo, Japan: IEEE Computer Society. Davis , F. D. ( 1989 ). Perceived usefulness, perceived ease of use and user acceptance of information technology . MIS Quarterly , 13 ( 3 ), 319 – 339 . Google Scholar Crossref Search ADS WorldCat Davis , F. D. , Bagozzi , R. P., & Warshaw , P. R. ( 1989 ). User acceptance of computer technology: A comparison of two theoretical models . Management Science , 35 , 982 – 1003 . Google Scholar Crossref Search ADS WorldCat Dowling , G. , & Staelin , R. ( 1994 ). A model of perceived risk and intended risk-handling activity . Journal of Consumer Research , 21 , 119 – 134 . Google Scholar Crossref Search ADS WorldCat Emirbayerand , M. , & Goodwin , J. ( 1994 ). Network analysis, culture, and the problem of agency . American Journal of Sociology , 99 ( 6 ), 1411 – 1454 . Google Scholar Crossref Search ADS WorldCat Featherman , M. ( 2001 ). Extending the technology acceptance model by inclusion of perceived risk . AMCIS Proceedings , August 3–5, 2001, Boston, MA: Association for Information Systems. Featherman , M. S. , & Pavlou , P. A. ( 2003 ). Predicting e-services adoption: A perceived risk facets perspective . International of Human–Computer Studies , 59 ( 4 ), 451 – 474 . Google Scholar Crossref Search ADS WorldCat Fishbein , M. , & Ajzen , I. ( 1975 ). Belief, attitude, intention, and behavior: An introduction to theory and research . Reading, MA, U.S.A. : Addison-Wesley Longman Publishing Co., Inc . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Freebersyser , J. A. , & Leiner , B. ( 2001 ). A DoD perspective on mobile ad hoc networks . In Charles E. Perkins (Ed.), Ad hoc networking (pp. 29 – 51 ). Boston, MA, U.S.A. : Addison Wesley Longman Publishing Co., Inc . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Germunden , H. G. ( 1985 ). Perceived risk and information search: A systematic meta-analysis of empirical evidence . International Journal of Research in Marketing , 2 , 79 – 100 . Google Scholar Crossref Search ADS WorldCat Goldsmith , A. J. , & Wicker , S. B. ( 2002 ). Design challenges for energy-constrained ad hoc wireless networks . IEEE Wireless Communications , 9 ( 4 ), 8 – 27 . Google Scholar Crossref Search ADS WorldCat Hoffman , D. , Novak , T., & Peralta , M. ( 1999 ). Building consumer trust online . Communications of the ACM , 42 ( 4 ), 80 – 85 . Google Scholar Crossref Search ADS WorldCat Hogg , M. A. ( 1992 ). The social psychology of group cohesiveness: From attraction to social identity . New York : New York University Press . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Hong , X. , Kong , J., & Gerla , M. ( 2006 ). Mobility changes anonymity: New passive threats in mobile ad hoc networks . Wireless Communications and Mobile Computing , 6 ( 3 ), 281 – 293 . Google Scholar Crossref Search ADS WorldCat Hsu C.-L. , & Lu , H.-P. ( 2007 ). Consumer behavior in online game communities: A motivational factor perspective . Computers in Human Behavior , 23 ( 3 ), 1642 – 1659 . Google Scholar Crossref Search ADS WorldCat Ilyas , M. ( 2003 ). The handbook of ad hoc mobile wireless networks . Boca Raton, FL, U.S.A., CRC Press . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Kahneman , D. , & Tversky , A. ( 1979 ). Prospect theory: An analysis of decision under risk . Econometrica , 47 ( 2 ), 263 – 291 . Google Scholar Crossref Search ADS WorldCat Khalifa , M. , & Cheng , S. K. N. ( 2002 ). Adoption of mobile commerce: Role of exposure. 35thHawaii International Conference on System Sciences, January 7–10, 2002, Hilton Waikoloa Village, Island of Hawaii: University of Hawai'i College of Business Administration . Retrieved from http://csdl2.computer.org/comp/proceedings/hicss/2002/1435/01/14350046.pdf. Kim , D. J. ( 2005 ). Cognition-based versus affect-based trust determinations in e-commerce: A cross-cultural comparison study . 26th International Conference on Information Systems, December 11–14, 2005, Las Vegas U.S.A.: Association for Information Systems . Kim , H.-W. , Chan , H. C., & Gupta , S. ( 2007 ). Value-based adoption of mobile internet: An empirical investigation . Decision Support Systems , 43 ( 1 ), 111 – 126 . Google Scholar Crossref Search ADS WorldCat Klein , K. J. , & Sorra , J. S. ( 1996 ). The challenge of innovation implementation . The Academy of Management Review , 21 ( 4 ), 1055 – 1080 . Google Scholar OpenURL Placeholder Text WorldCat Koivisto , M. , & Urbaczewski , A. ( 2004 ). The relationship between quality of service perceived and delivered in mobile Internet communications . Information Systems and E-Business Management , 2 ( 4 ), 309 – 323 . Google Scholar Crossref Search ADS WorldCat Kuisma , T. , Laukkanen , T., & Hiltunen , M. ( 2007 ). Mapping the reasons for resistance to Internet banking: A means–end approach . International Journal of Information Management , 27 ( 2 ), 75 – 85 . Google Scholar Crossref Search ADS WorldCat Li , H. , Weckerle , M., Zirwas , W., & Schulz , E. ( 2002 ). Multihop communications in future mobile radio networks . 13th IEEE International Symposium on Personal Indoor and Mobile Radio Communications, September 15–18, 2002, Lisbon, Portugal: Instituto Superior Técnico . Lindell , M. K. , & Whiteny , D. J. ( 2001 ). Accounting for common method variance in cross-sectional research designs . Journal of Applied Psychology , 86 ( 1 ), 114 – 121 . Google Scholar Crossref Search ADS PubMed WorldCat Lu , J. , Yu , C.-S., Liu , C., & Yao , J. E. ( 2003 ). Technology acceptance model for wireless Internet . Electronic Networking Applications and Policy , 13 ( 3 ), 206 – 222 . Google Scholar Crossref Search ADS WorldCat Maeshima , O. , Fuke , N., Sugiyama , K., Shinonaga , H., & Acampora , A. ( 2003 ). Simulation evaluation on area extension and system capacity of CDMA packet communication system utilizing multihop connection . 14th Personal, Indoor and Mobile Radio Communications, September 7–10, 2003, Beijing, China: IEEE . Mason , R. ( 1986 ). Four ethical issues of the information age . MIS Quarterly , 10 ( 1 ), 4 – 12 . Google Scholar Crossref Search ADS WorldCat Mukherji , A. , Wright , P., & Mukherji , J. ( 2007 ). Cohesiveness and goals in agency networks: Explaining conflict and cooperation . Journal of Socio-Economics , 36 ( 6 ), 949 – 964 . Google Scholar Crossref Search ADS WorldCat Mitchell , V. W. ( 1992 ). Understanding consumers’ behavior: Can perceived risk theory help? Management Decision , 30 ( 2 ), 26 – 31 . Google Scholar Crossref Search ADS WorldCat Nguyen , D.-Q. , & Minet , P. ( 2006 ). QoS support and OLSR routing in a mobile ad hoc network . International Conference on Systems and International Conference on Mobile Communications and Learning Technologies , April 23–29, 2006, Washington, DC, U.S.A.: IEEE Computer Society . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Pavlou , P. A. ( 2001 ). Integrating trust in electronic commerce with the technology acceptance model: Model development and validation . AMCIS Proceedings , August 3–5, 2001, Boston, Massachusetts, U.S.A.: Association for Information Systems . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Pavlou , P. A. , Liang , H., & Xue , Y. ( 2007 ). Understanding and mitigating uncertainty in online exchange relationship: A principal–agent perspective . MIS Quarterly , 31 ( 1 ), 105 – 136 . Google Scholar OpenURL Placeholder Text WorldCat Peter , J. P. , & Ryan , M. J. ( 1976 ). An investigation of perceived risk at the brand level . Journal of Marketing Research , 13 ( 2 ), 184 – 188 . Google Scholar Crossref Search ADS WorldCat RACE . ( 1994 ). UMTS system structure document, Issue 1.0. RACE 2066 Mobile network (MONET) , CEC Deliverable No:R2066/LMF/GA1/DS/P/052/b1. Ram , S. ( 1987 ). A model of innovation resistance . Advances in Consumer Research , 14 ( 1 ), 208 – 212 . Google Scholar OpenURL Placeholder Text WorldCat Rogers , E. M. ( 1995 ). Diffusion of Innovation (4th ed.). New York : Free Press . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Salem , N. B. , Hubaux , J., & Jakobsson , M. ( 2004 ). Reputation-based Wi-Fi deployment—Protocols and security Analysis . Wireless Mobile Applications and Services on WLAN Hotspots , October 1, 2004. Philadelphia, PA, U.S.A. : ACM . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Saliba , A. J. , Beresford , M. A., Invanovich , M, & Fitzpatrick , P. ( 2005 ). User-perceived quality of service in wireless data networks . Personal and Ubiquitous Computing , 9 ( 6 ), 413 – 422 . Google Scholar Crossref Search ADS WorldCat Shelper , J. ( 2005 ). 1G, 2G, 3G, 4G . TechColumn , April 2005. Retrieved July 28, 2008 from http://www.T1Rex.com. Sheth , J. N. ( 1981 ). Psychology of innovation resistance: The less developed concept (LDC) in diffusion research . Research in Marketing , 4 ( 3 ), 273 – 282 . Google Scholar OpenURL Placeholder Text WorldCat Smith , H. J. , Milberg , S., & Burke , S. ( 1996 ). Information privacy: Measuring individuals’ concerns about organizational practices . MIS Quarterly , 20 ( 2 ), 167 – 196 . Google Scholar Crossref Search ADS WorldCat Suliman , I. M. , Oppermann , I., Braysy , T., Konnov , I., & Laitinen , E. ( 2005 ). A cooperative multihop radio resource allocation in next generation networks . IEEE Vehicular Technology Conference , 61 ( 4 ), 2400 – 2404 . Google Scholar OpenURL Placeholder Text WorldCat Szmigin , I. , & Foxall , G. ( 1998 ). Three forms of innovation resistance: The case of retail payment method . Technovation , 18 ( 6/7 ), 459 – 468 . Google Scholar Crossref Search ADS WorldCat Tobia , R. D. ( 1996 ). An introduction to partial least squares regression . SAS Institution Inc ., Cary, NC. Retrieved from http://www.ats.ucla.edu/stat/sas/library/pls.pdf. Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Venkatesh , V. , Morris , M. G., Davis , G. B., & Davis , F. D. ( 2003 ). User acceptance of information technology: Toward a unified view . MIS Quarterly , 27 ( 3 ), 425 – 478 . Google Scholar OpenURL Placeholder Text WorldCat Wold , H. ( 1985 ). Partial least squares . In S. Kotz and N. Johnson (Eds.), Encyclopedia of statistical sciences (pp. 58 – 591 ). New York : Wiley . Google Scholar Google Preview OpenURL Placeholder Text WorldCat COPAC Zhou , Y. ( 2008 ). Voluntary adopters versus forced adopters: Integrating the diffusion of innovation theory and the technology acceptance model to study intra-organizational adoption . New Media & Society , 10 ( 3 ), 475 – 496 . Google Scholar Crossref Search ADS WorldCat About the Authors Yoolee Kang (xiaojie622@kisdi.re.kr) is a researcher in Fair Competition Policy Division at Korea Information Society Development Institute. Her research focuses on telecommunication user empowerment and fair competition policy in broadcasting and telecommunications area. Address: Fair Competition Policy Division, Korea Information Society Development Institute, 1-1 Juam-Dong, Gwacheon-Si, Gyeonggi-Do, 427-710, Korea. Seongcheol Kim (hiddentrees@korea.ac.kr) is currently an associate professor in the School of Journalism and Mass Communication at Korea University. He received his M.A. and Ph.D. degrees in telecommunications from Michigan State University. Before joining Korea University, he led e-government initiatives for the Seoul Metropolitan Government as information systems director and was an associate professor in the School of IT Business at Information and Communications University. His research interests include telecommunications management, mobile media, and convergence. Address: School of Journalism and Mass Communication, Korea University, Anam-Dong, Seongbuk-Gu, Seoul 136-701, Korea. Appendix Table A Measurement items . Item . Source . ENQC1 Without a fixed route, the service may not be available when necessary, because mobile users move around. Nguyen & Minet, 2006 ENQC2 Services will not be provided efficiently, as user mobility will change the route of service transmission. ENQC3 It is hard to predict network quality as mobile devices of other users move. ENQC4 I will not be satisfied with the services transmitted by other mobile users’ devices. EPC1 It bothers me that my location information may be exposed by passing through other users’ mobile devices. Pavlou et al., 2007; Smith et al., 1996 EPC2 I am concerned that other users are collecting too much information (e.g., location information) about me. EPC3 Those who pass through my personal mobile device could access information (e.g., my calling history) stored in it. EPC4 My personal information could be retained in others’ mobile devices during multihop communications. ELOC1 Those who use multihop communications tend to be unrelated to me either personally or socially. Hsu & Lu, 2007 ELOC2 Unlike current communications in which only beneficiaries participate, multihop communications may create a less cohesive atmosphere. ELOC3 I think that places that I belong to are more suitable for multihop communications than public places where I feel less sense of belonging. ERS1 I do not like to consume my resources (battery or memory) in order to deliver services via my personal mobile device. Bagozzi & Dholakia, 2006 ERS2 I will turn off my personal mobile devices for future use if resources (battery or memory) are almost out. ERS3 I am worried that I will not be able to use my personal mobile device because of resource (battery or memory) shortages when other users communicate through it. RES1 I feel uneasy when data are transmitted by multihop communications. Klein & Sorra, 1996 RES2 I insist upon using services delivered through fixed infrastructure (base stations or access points) rather than via multihop communications. RES3 Multihop communications via users’ mobile devices deserve criticism. RES4 I am dissatisfied with multihop communications via users’ mobile devices. EU1 Multihop communications enhance my satisfaction in daily life. Davis, 1989 EU2 Multihop communications services (e.g., vehicular applications, buddy finding) will help me to achieve what I want to do easily. EU3 Multihop communications services (e.g., vehicular applications, buddy finding) will reduce time and costs. EEOP1 It is easy to use multihop services as an end user. Davis, 1989 EEOP2 It is simple and clear to go through others’ mobile devices in order to use services that I want. EEOP3 It is easy to route others’ data using multihop communications as intermediary user. PI1 I plan to participate as an intermediary route in multihop communications using my mobile device. Davis, 1989 PI2 I plan to use services delivered by others’ mobile devices. PI3 I will advise others to participate in multihop communications. . Item . Source . ENQC1 Without a fixed route, the service may not be available when necessary, because mobile users move around. Nguyen & Minet, 2006 ENQC2 Services will not be provided efficiently, as user mobility will change the route of service transmission. ENQC3 It is hard to predict network quality as mobile devices of other users move. ENQC4 I will not be satisfied with the services transmitted by other mobile users’ devices. EPC1 It bothers me that my location information may be exposed by passing through other users’ mobile devices. Pavlou et al., 2007; Smith et al., 1996 EPC2 I am concerned that other users are collecting too much information (e.g., location information) about me. EPC3 Those who pass through my personal mobile device could access information (e.g., my calling history) stored in it. EPC4 My personal information could be retained in others’ mobile devices during multihop communications. ELOC1 Those who use multihop communications tend to be unrelated to me either personally or socially. Hsu & Lu, 2007 ELOC2 Unlike current communications in which only beneficiaries participate, multihop communications may create a less cohesive atmosphere. ELOC3 I think that places that I belong to are more suitable for multihop communications than public places where I feel less sense of belonging. ERS1 I do not like to consume my resources (battery or memory) in order to deliver services via my personal mobile device. Bagozzi & Dholakia, 2006 ERS2 I will turn off my personal mobile devices for future use if resources (battery or memory) are almost out. ERS3 I am worried that I will not be able to use my personal mobile device because of resource (battery or memory) shortages when other users communicate through it. RES1 I feel uneasy when data are transmitted by multihop communications. Klein & Sorra, 1996 RES2 I insist upon using services delivered through fixed infrastructure (base stations or access points) rather than via multihop communications. RES3 Multihop communications via users’ mobile devices deserve criticism. RES4 I am dissatisfied with multihop communications via users’ mobile devices. EU1 Multihop communications enhance my satisfaction in daily life. Davis, 1989 EU2 Multihop communications services (e.g., vehicular applications, buddy finding) will help me to achieve what I want to do easily. EU3 Multihop communications services (e.g., vehicular applications, buddy finding) will reduce time and costs. EEOP1 It is easy to use multihop services as an end user. Davis, 1989 EEOP2 It is simple and clear to go through others’ mobile devices in order to use services that I want. EEOP3 It is easy to route others’ data using multihop communications as intermediary user. PI1 I plan to participate as an intermediary route in multihop communications using my mobile device. Davis, 1989 PI2 I plan to use services delivered by others’ mobile devices. PI3 I will advise others to participate in multihop communications. Open in new tab Table A Measurement items . Item . Source . ENQC1 Without a fixed route, the service may not be available when necessary, because mobile users move around. Nguyen & Minet, 2006 ENQC2 Services will not be provided efficiently, as user mobility will change the route of service transmission. ENQC3 It is hard to predict network quality as mobile devices of other users move. ENQC4 I will not be satisfied with the services transmitted by other mobile users’ devices. EPC1 It bothers me that my location information may be exposed by passing through other users’ mobile devices. Pavlou et al., 2007; Smith et al., 1996 EPC2 I am concerned that other users are collecting too much information (e.g., location information) about me. EPC3 Those who pass through my personal mobile device could access information (e.g., my calling history) stored in it. EPC4 My personal information could be retained in others’ mobile devices during multihop communications. ELOC1 Those who use multihop communications tend to be unrelated to me either personally or socially. Hsu & Lu, 2007 ELOC2 Unlike current communications in which only beneficiaries participate, multihop communications may create a less cohesive atmosphere. ELOC3 I think that places that I belong to are more suitable for multihop communications than public places where I feel less sense of belonging. ERS1 I do not like to consume my resources (battery or memory) in order to deliver services via my personal mobile device. Bagozzi & Dholakia, 2006 ERS2 I will turn off my personal mobile devices for future use if resources (battery or memory) are almost out. ERS3 I am worried that I will not be able to use my personal mobile device because of resource (battery or memory) shortages when other users communicate through it. RES1 I feel uneasy when data are transmitted by multihop communications. Klein & Sorra, 1996 RES2 I insist upon using services delivered through fixed infrastructure (base stations or access points) rather than via multihop communications. RES3 Multihop communications via users’ mobile devices deserve criticism. RES4 I am dissatisfied with multihop communications via users’ mobile devices. EU1 Multihop communications enhance my satisfaction in daily life. Davis, 1989 EU2 Multihop communications services (e.g., vehicular applications, buddy finding) will help me to achieve what I want to do easily. EU3 Multihop communications services (e.g., vehicular applications, buddy finding) will reduce time and costs. EEOP1 It is easy to use multihop services as an end user. Davis, 1989 EEOP2 It is simple and clear to go through others’ mobile devices in order to use services that I want. EEOP3 It is easy to route others’ data using multihop communications as intermediary user. PI1 I plan to participate as an intermediary route in multihop communications using my mobile device. Davis, 1989 PI2 I plan to use services delivered by others’ mobile devices. PI3 I will advise others to participate in multihop communications. . Item . Source . ENQC1 Without a fixed route, the service may not be available when necessary, because mobile users move around. Nguyen & Minet, 2006 ENQC2 Services will not be provided efficiently, as user mobility will change the route of service transmission. ENQC3 It is hard to predict network quality as mobile devices of other users move. ENQC4 I will not be satisfied with the services transmitted by other mobile users’ devices. EPC1 It bothers me that my location information may be exposed by passing through other users’ mobile devices. Pavlou et al., 2007; Smith et al., 1996 EPC2 I am concerned that other users are collecting too much information (e.g., location information) about me. EPC3 Those who pass through my personal mobile device could access information (e.g., my calling history) stored in it. EPC4 My personal information could be retained in others’ mobile devices during multihop communications. ELOC1 Those who use multihop communications tend to be unrelated to me either personally or socially. Hsu & Lu, 2007 ELOC2 Unlike current communications in which only beneficiaries participate, multihop communications may create a less cohesive atmosphere. ELOC3 I think that places that I belong to are more suitable for multihop communications than public places where I feel less sense of belonging. ERS1 I do not like to consume my resources (battery or memory) in order to deliver services via my personal mobile device. Bagozzi & Dholakia, 2006 ERS2 I will turn off my personal mobile devices for future use if resources (battery or memory) are almost out. ERS3 I am worried that I will not be able to use my personal mobile device because of resource (battery or memory) shortages when other users communicate through it. RES1 I feel uneasy when data are transmitted by multihop communications. Klein & Sorra, 1996 RES2 I insist upon using services delivered through fixed infrastructure (base stations or access points) rather than via multihop communications. RES3 Multihop communications via users’ mobile devices deserve criticism. RES4 I am dissatisfied with multihop communications via users’ mobile devices. EU1 Multihop communications enhance my satisfaction in daily life. Davis, 1989 EU2 Multihop communications services (e.g., vehicular applications, buddy finding) will help me to achieve what I want to do easily. EU3 Multihop communications services (e.g., vehicular applications, buddy finding) will reduce time and costs. EEOP1 It is easy to use multihop services as an end user. Davis, 1989 EEOP2 It is simple and clear to go through others’ mobile devices in order to use services that I want. EEOP3 It is easy to route others’ data using multihop communications as intermediary user. PI1 I plan to participate as an intermediary route in multihop communications using my mobile device. Davis, 1989 PI2 I plan to use services delivered by others’ mobile devices. PI3 I will advise others to participate in multihop communications. Open in new tab Table B Individual items’ factor loading, construct CR, and AVE Measurement items . Loading . CR . AVE . Expected network quality concerns ENQC1 .7274 .893 .678 ENQC 2 .8641 ENQC 3 .8473 ENQC 4 .8461 Expected privacy concerns EPC1 .8860 .944 .807 EPC2 .9057 EPC3 .9060 EPC4 .8952 Expected lack of cohesion ELOC1 .8363 .859 .671 ELOC2 .8707 ELOC3 .7453 Expected resource sacrifice ERS1 .8688 .901 .753 ERS2 .8412 ERS3 .8926 Resistance RES1 .8362 .906 .706 RES2 .8204 RES3 .8219 RES4 .8813 Expected usefulness EU1 .8648 .902 .755 EU2 .8854 EU3 .8558 Expected ease of participation EEOP1 .8627 .911 .773 EEOP2 .9098 EEOP3 .8648 Participation intention PI1 .9169 .943 .847 PI2 .9286 PI3 .9146 Measurement items . Loading . CR . AVE . Expected network quality concerns ENQC1 .7274 .893 .678 ENQC 2 .8641 ENQC 3 .8473 ENQC 4 .8461 Expected privacy concerns EPC1 .8860 .944 .807 EPC2 .9057 EPC3 .9060 EPC4 .8952 Expected lack of cohesion ELOC1 .8363 .859 .671 ELOC2 .8707 ELOC3 .7453 Expected resource sacrifice ERS1 .8688 .901 .753 ERS2 .8412 ERS3 .8926 Resistance RES1 .8362 .906 .706 RES2 .8204 RES3 .8219 RES4 .8813 Expected usefulness EU1 .8648 .902 .755 EU2 .8854 EU3 .8558 Expected ease of participation EEOP1 .8627 .911 .773 EEOP2 .9098 EEOP3 .8648 Participation intention PI1 .9169 .943 .847 PI2 .9286 PI3 .9146 Open in new tab Table B Individual items’ factor loading, construct CR, and AVE Measurement items . Loading . CR . AVE . Expected network quality concerns ENQC1 .7274 .893 .678 ENQC 2 .8641 ENQC 3 .8473 ENQC 4 .8461 Expected privacy concerns EPC1 .8860 .944 .807 EPC2 .9057 EPC3 .9060 EPC4 .8952 Expected lack of cohesion ELOC1 .8363 .859 .671 ELOC2 .8707 ELOC3 .7453 Expected resource sacrifice ERS1 .8688 .901 .753 ERS2 .8412 ERS3 .8926 Resistance RES1 .8362 .906 .706 RES2 .8204 RES3 .8219 RES4 .8813 Expected usefulness EU1 .8648 .902 .755 EU2 .8854 EU3 .8558 Expected ease of participation EEOP1 .8627 .911 .773 EEOP2 .9098 EEOP3 .8648 Participation intention PI1 .9169 .943 .847 PI2 .9286 PI3 .9146 Measurement items . Loading . CR . AVE . Expected network quality concerns ENQC1 .7274 .893 .678 ENQC 2 .8641 ENQC 3 .8473 ENQC 4 .8461 Expected privacy concerns EPC1 .8860 .944 .807 EPC2 .9057 EPC3 .9060 EPC4 .8952 Expected lack of cohesion ELOC1 .8363 .859 .671 ELOC2 .8707 ELOC3 .7453 Expected resource sacrifice ERS1 .8688 .901 .753 ERS2 .8412 ERS3 .8926 Resistance RES1 .8362 .906 .706 RES2 .8204 RES3 .8219 RES4 .8813 Expected usefulness EU1 .8648 .902 .755 EU2 .8854 EU3 .8558 Expected ease of participation EEOP1 .8627 .911 .773 EEOP2 .9098 EEOP3 .8648 Participation intention PI1 .9169 .943 .847 PI2 .9286 PI3 .9146 Open in new tab Table C Cross loading for measurement model Open in new tab Table C Cross loading for measurement model Open in new tab Table D Correlations of the latent variables and square root of the AVE Latent variables . (1) . (2) . (3) . (4) . (5) . (6) . (7) . (8) . (1) Expected network quality concerns (.823) (2) Expected privacy concerns .585 (.898) (3) Expected lack of cohesion .557 .644 (.820) (4) Expected resource sacrifice .520 .638 .592 (.868) (5) Resistance .478 .421 .447 .460 (.841) (6) Expected usefulness -.004 .120 .155 .083 -.123 (.869) (7) Expected ease of participation -.042 .057 .130 .000 .001 .538 (.880) (8) Participation intention -.166 -.182 -.075 -.197 -.281 .450 .642 (.920) Latent variables . (1) . (2) . (3) . (4) . (5) . (6) . (7) . (8) . (1) Expected network quality concerns (.823) (2) Expected privacy concerns .585 (.898) (3) Expected lack of cohesion .557 .644 (.820) (4) Expected resource sacrifice .520 .638 .592 (.868) (5) Resistance .478 .421 .447 .460 (.841) (6) Expected usefulness -.004 .120 .155 .083 -.123 (.869) (7) Expected ease of participation -.042 .057 .130 .000 .001 .538 (.880) (8) Participation intention -.166 -.182 -.075 -.197 -.281 .450 .642 (.920) Open in new tab Table D Correlations of the latent variables and square root of the AVE Latent variables . (1) . (2) . (3) . (4) . (5) . (6) . (7) . (8) . (1) Expected network quality concerns (.823) (2) Expected privacy concerns .585 (.898) (3) Expected lack of cohesion .557 .644 (.820) (4) Expected resource sacrifice .520 .638 .592 (.868) (5) Resistance .478 .421 .447 .460 (.841) (6) Expected usefulness -.004 .120 .155 .083 -.123 (.869) (7) Expected ease of participation -.042 .057 .130 .000 .001 .538 (.880) (8) Participation intention -.166 -.182 -.075 -.197 -.281 .450 .642 (.920) Latent variables . (1) . (2) . (3) . (4) . (5) . (6) . (7) . (8) . (1) Expected network quality concerns (.823) (2) Expected privacy concerns .585 (.898) (3) Expected lack of cohesion .557 .644 (.820) (4) Expected resource sacrifice .520 .638 .592 (.868) (5) Resistance .478 .421 .447 .460 (.841) (6) Expected usefulness -.004 .120 .155 .083 -.123 (.869) (7) Expected ease of participation -.042 .057 .130 .000 .001 .538 (.880) (8) Participation intention -.166 -.182 -.075 -.197 -.281 .450 .642 (.920) Open in new tab Copyright © 2009 International Communication Association
TI - Understanding User Resistance to Participation in Multihop Communications
JF - Journal of Computer-Mediated Communication
DO - 10.1111/j.1083-6101.2009.01443.x
DA - 2009-01-01
UR - https://www.deepdyve.com/lp/oxford-university-press/understanding-user-resistance-to-participation-in-multihop-YJ6XkgrN06
SP - 328
EP - 351
VL - 14
IS - 2
DP - DeepDyve
ER -