• 沒有找到結果。

Understanding business-level innovation technology adoption

N/A
N/A
Protected

Academic year: 2021

Share "Understanding business-level innovation technology adoption"

Copied!
37
0
0

加載中.... (立即查看全文)

全文

(1)

Yu, J.-S. and Tao, Y.-H., Understanding business-level innovation technology adoption,

Technovation, Vol. 29, No. 2, 2009, 92-109

Understanding Business-level innovation technology adoption

Abstract

The implementation of new Internet-based information system and technology (IT/IS) has been recognized as an important process for transforming a business toward electronic business. In line with this perspective, the business attitudes regarding the adoption of innovation IT/IS have been recognized as a critical factor for executing electronic business strategy. Since extant studies attempting to find influences on the individual adoption of IT/IS are dominated by technology acceptance model (TAM), this study attempts to extend TAM to business-level innovation technology adoption. Empirical results indicate that perceived usefulness, subject norm, perceived easy-of-use, and characteristics of the firm itself are very important factors influencing attitudes of businesses at the pre-decision stage, while only perceived usefulness and subject norm significantly affect attitudes of businesses at the in-decision stage. Additionally, the effect of perceived easy-of-use on both perceived usefulness and company attitudes as well as the influence of perceived usefulness on firm attitude are changeable, and rely on the complexity of the innovation IT/IS itself. The theoretical and business implications are discussed.

Keywords: Technology acceptance model; Innovation diffusion theory; Electronic business; Electronic marketplace

1. Introduction

Accelerated growth of the Internet and electronic commerce (e-commerce) in the recent decade has forced businesses to encounter global competition and encouraged them to establish a presence in global markets via the implementation of new Internet-based information system and technology (IT/IS). Given that the implementation of new Internet-based IT/IS is a continued adoption process for transforming a business toward electronic business (e-business), establishing electronic links with its suppliers and buyers, and executing electronic transactions along value-chain activities, the business attitudes regarding the adoption of innovation IT/IS have been recognized as a critical factor for executing e-business strategy. However, e-business is different from previous traditional technological innovation (Lin and Lin,

(2)

2008). In contrast, e-business represents a new innovative approach to incorporate core business processes/functions with Internet-based IT/IS (Zhu, 2004; Teo et al., 2006; Lin and Lin, 2008).

Current studies attempting to find the determinants influencing individual-level IT/IS adoption are heavily based on behavioral theories such as technology acceptance model (TAM), theory of planned behavior (TPB), and innovation diffusion theory (IDT) (Hernandez et al., 2008). Literature on business-level technology adoption is scarce compared to general literature on examining individual-level technology adoption, and in particular contains few studies adopting the TAM standpoint. Enterprises allocate significant portions of their budget each year to procuring new IT/IS, and this trend has become more obvious following the advance of IT/IS and the diffusion and development of the e-life, e-society, and e-business. Hence, understanding business-level innovative technology adoption is just as important as understanding individual-level new technology adoption.

This study chooses electronic marketplace (e-marketplace) as the study object, because the e-marketplace is an Internet-based IT/IS innovation and application. Although rapid growth of e-marketplaces appears inevitable, a survey undertaken in early 2004 (Yu, 2006) indicated that the adoption rate of e-marketplaces by Taiwanese enterprises was approximately 23.48%, which was well below expectations. Therefore, the reasons why certain enterprises wish to use the e-marketplaces, while others do not, are quite interesting to be investigated. Compared to the vast existing literature on e-marketplaces, relatively few works have studied business to business (B2B) e-marketplaces adoption from behavioral theories, and most such studies are conducted from the economic viewpoint (Bakos, 1991, 1997; Strader & Shaw, 1999; Benslimane et al., 2005; Zhu et a., 2006) which may not fully explain the B2B e-marketplace adoption (Driedonks et al., 2005).

Motivated by the above discussion, this study focuses on three objectives. First, this work examines how business-level attitudes influence innovative Internet-based IT/IS adoption. Second, in contrast to the previous TAM-based literature, which generally takes individual-level users as the survey unit, this study takes collective organizations as the analysis unit to examine whether TAM remain valid at the business-level technology adoption. Third, rather than using an economic perspective, and only providing a static view in examining the influences on e-marketplace adoption by enterprises, this paper integrates TAM and IDT to form a multi-model research structure that reveals a dynamic picture of on a firm’s attitudes before new technology adoption,

(3)

decisions to adopt the new technology, and decisions to continue using or rejecting it. 2. Theoretical framework

Compared to the large body of individual-level TAM literature, business-level TAM literature is relatively rare, though studies using TAM to examine organizational-level technology adoption are not entirely novel (Amoako-Gyampah and Salam, 2004; Zain et al., 2005). However, such researches have failed to clarify the relationship between business-level attitude and behavior on innovative technology adoption. This implies that the underlying technology adoption at the firm level has not been discussed and ascertained in sufficient detail.

2.1 TAM

TAM, proposed by Davis in 1986 (Davis, 1989), is used to effectively forecast individual computer acceptance behavior, and was adapted from theory of reasoned action (TRA) developed by Ajzen and Fishbein in 1975. In TAM, the actual behavior (AB) of an individual to adopt a technology-based product can be predicted by the perceived usefulness (PU) and perceived ease-of-use (PEOU) of that individual as expressed by the regression model AB = 1PU+2PEOU+ε, where PU is defined as

the subjective assessment of a user or prospective user that using the product will provide benefits related to job performance, and PEOU is the degree to which an individual can use the product free of effort (Davis et al., 1989).

Since a business comprises a group of individuals, meaning business behavior is collective behavior of individuals, the usefulness of business-level TAM can be defined as the number of benefits obtainable by the company using the new technology, which is subjectively evaluated by key decision makers in firms. Likewise, ease-of- use can be defined as the degree to which business can effortlessly use the new technology. Effort in this context can refer to monetary investment, employee training time, technology switching barriers, maintenance costs, and so on.

Over the past two decades, enormous studies have used TAM or related extensions to provide empirical evidence on the relationships among PU, PEOU and AB, or to validate and enhance the reliability and robustness of the TAM questionnaire instrument. Notably, although Davis et al. (1989) argued that the subjective norm (SN) did not significantly influence usage intention, and thus omitted SN in their original TAM, Davis modified this approach (Venkatesh & Davis, 2000) and concluded that SN

(4)

considerably influences the attitude toward IT product adoption, based on numerous empirical studies demonstrating this (Hartwick & Barki, 1994; Karahanna et al., 1999). From an organizational behavior perspectives, many studies found that organizational decision behavior not only inherits the rational and irrational components of individual decisions but is also a collective perception reflecting the concerns of multi-dimensional stakeholders (Frambach and Schillewaert, 2002; Nelson & Quick, 2006).

Building on the above discussion, the new technology adoption behavior demonstrated by the whole business might resemble that demonstrated by a single individual. Accordingly, generalized business-level technology adoption attitude and behavior maybe can also be effectively explained by TAM, as shown in Fig. 1. Since this study explores the influences of firm attitude, decision, and continuance on e-marketplace adoption/non-adoption, the term “consumer”as used in the remainder of this paper may refer to a business, firm, or organization.

Insert Fig. 1 here.

2.2 IDT

IDT, pioneered by Rogers in 1962 (Rogers, 2003), is used as a process-oriented perspective to explain how an innovation can be accepted and disseminated among consumers. IDT contends that the adoption or rejection of an innovation begins with consumer awareness of that innovation, while diffusion is a process in which information regarding an innovation is conveyed via certain channels over time among consumers. Innovation is defined as an idea, practice, product, or object that consumers perceive as new. Despite receiving numerous positive and negative criticisms corresponding to different cognitive styles, IDT has been applied to over thousands of empirical studies since 1962 (Rogers, 2003) including studies of business-level innovation technology adoption (Gatignon & Robertson, 1989; Cooper & Zmud, 1990; Chau & Tam, 2000; Frambach & Schillewaert, 2002). Because many innovations require a lengthy period before being popular, which frequently lasts for many years, IDT is quite useful in identifying ways to accelerate innovation diffusion from the time they become available to the time they are widely adopted (Rogers, 2003).

IDT maintains that, during the pre-decision stage, consumers actively seek and/or passively receive information regarding innovations and shape their favorable or unfavorable beliefs regarding the innovations. The in-decision stage takes place when a

(5)

consumer engages in activities that lead to them making a choice between adopting and rejecting an innovation. Meanwhile, the post-decision stage occurs immediately after consumers begin using or reject using an innovation. During the post-decision stage, consumers seek reinforcement for their previous decision, and may reverse the decision if exposed to related dissonance messages. Restated, non-adopters may either continuously reject the use of the innovation or choose to adopt it. Non-adopters adopt an innovation if motivated to do so after securing further information or evidence that influences their original beliefs of not adopting the innovation. Conversely, adopters may continuously use the innovation or alternatively may stop using it owing to dissatisfaction with its performance. Figure 2 shows the IDT process of Rogers.

Insert Fig. 2 here.

3. Research Structure

E-marketplace development has only just begun its steady growth during the recent years, and represents an innovation Internet-based IT, IS, or business model. However, the initial idea of establishing a cybernetic buying and selling platform was originally presented by several authors during the period long before the inception of the Internet or e-commerce (McFarlan, 1984; Malone et al., 1987; Bakos, 1991). Despite the existence of a large body of e-marketplace studies and literature regarding B2B e-marketplace adoption dating back to the early 1990s (Bakos, 1991; Lee & Clark, 1996), literature directly related to firm-level e-marketplaces is relatively scarce and is briefly summarized in Table 1.

Take in Table 1

Instead of describing the development of the research model, the paper presents the formulated research model based on the research objectives setup in the “Introduction” section, and the principles established next in this section. Also, the research hypotheses are derived with the research model concurrently, followed by the justification of the determinants proposed in the research model and hypotheses.

3.1 Principles of structure formulation

The beauty of Davis’TAM is simple and effective by employing two simple constructs, PU and PEOU, to replace many of TRA constructs while being able to explain the adoption behavior from the perspective of either the critical reviews

(6)

(Bogozzi, 2007; Benbasat & Barki, 2007) or meta analyses (King & He 2006; Yousafzai et al., 2007). Because of this merit, there was plenty of room for many extended TAM studies since its inception, which according to Bogazzi (2007) had attracted an incredible number of over 700 citations. Similarly, as an initial study on firm-level TAM adoption behavior, the research model should not be too complicated, which not only provides a solid ground for extending studies but also prevent this model to be trivial in its scope as concluded by some of the critical reviews about recent TAM-related studies (Venkatesh et al, 2007; Bagozzi, 2007; Benbasat & Barki 2007). Accordingly, the first principle is to make the fundamental research structure as simple as possible for providing a solid research base for future firm-level TAM studies.

Comparing with extant e-marketplace literature heavily relying on an economic perspective and only providing a static view in examining the influences on e-marketplace adoption by enterprises, this paper attempts to integrate TAM and IDT to form a multi-model research structure that reveals a dynamic picture of on a firm’s attitudes before new technology adoption, decisions to adopt the new technology, and decisions to continue using or rejecting it. Therefore, the second principle is to observe firm-level adoption at different decision points derived from the TAM and IDT from a process-oriented perspective.

3.2 Research structure and hypotheses

First of all, grounded in TAM and IDT and based on the second principle of structure formulation, there are three models corresponding to the decision points, including pre-decision, in-decision, and post-decision, as depicted in Fig. 3. Compared to Davis’s TAM model (1989), this model further considered behavior at the adoption span and after the adoption, which provides a more precise distinguish of the critical factors influence the firm behavior from a process perspective. Accordingly, the hypotheses are divided into three sets at different decision points, and posited from TAM and the literature analysis of potential factors influencing the decision of B2B e-marketplace adoption which is discussed in subsection 3.3. A briefly rationalization behind the structure formulation is described in this subsection, and some model components are justified by the literature in next subsection.

Insert Fig. 3 here.

(7)

during the process, as also implied by the first principle of structure formulation. In Davis’s TAM model (1989), the antecedent factors influences three different-stage dependent variables, i.e., attitude, intention to adopt, and actual behavior, respectively. Accordingly, attitude is an observation before the adoption decision as seen in model 1, while the adoption decision itself is also influenced by the attitude as seen in model 2. Moreover, the behavior after decision may be different than the previous two, as seen in model 3, which according to Venkatesh & Davis (2007) is also an emergent research focus on continuous usage of IT/IS introduced by Bhatacherjee (2001). These different dependent variables can also be clearly seen in the three sets of hypotheses.

Notably, in order to observe the different dependent variables, all three models employee the same set of TAM constructs that will be justified in next subsection. Nevertheless, by applying the first principle of structure formulation, this study does not intend to exhaustedly cover all the constructs in previous TAM studies. Meanwhile, this study does not investigate their mutual relationships since it largely complicated the models. However, PEOU-PU relationship in pre-decision is included since it has examined by almost all individual-level TAM meta-analyses (Ma & Liu, 2004; King & He, 2006; Yousafzai et al, 2007; Schepers & Wetzels, 2007). In fact, the idea of using TAM to examine business-level technology adoption is not entirely new (Amoako-Gyampah & Salam, 2004; Zain et al., 2005). Based on a survey of 329 managers and executives in Malaysian manufacturing firms, Zain et al. (2005) found that PEOU not affects firm-level PU, while in a study of 1562 employees of American firms that focused on enterprise ERP adoption, Amoako-Gyampah and Salam (2004) concluded that firm-level PEOU strongly influences firm-level PU. Obviously, the finding between Zain et al. and Amoako-Gyampah and Salam is not consistent, which is worth an investigation in this firm-level TAM study.

Finally, the post-decision model distinguish adopters from non-adopters since it has to occur before the post-decision point of continue usage. That is why there are three hypotheses in model 3 compared to the similar model 2. It makes sense to understand those who did not adopt at the first place but changed their mind afterwards since valuable factors are hidden in this change behavior.

Built on the above, the hypotheses corresponding to these three models are posited as follows:

Hypotheses at the pre-decision stage

(8)

adoption.

H1a: PU significantly influences business attitude to e-marketplace adoption. H1b: PEOU significantly influences business attitude to e-marketplace adoption. H1c: SN significantly influences business attitude to e-marketplace adoption. H1d: FC significantly influences business attitude to e-marketplace adoption. H1e: ICEC significantly influences business attitude to e-marketplace adoption. H2: Business PEOU significantly influences its PU.

Hypotheses at the in-decision stage

H3: TAM constructs significantly influence business decision regarding e-marketplace adoption.

H3a: PU significantly influences business decision regarding e-marketplace adoption.

H3b: PEOU significantly influences business decision regarding e-marketplace adoption.

H3c: SN significantly influences business decision regarding e-marketplace adoption.

H3d: FC significantly influences business decision regarding e-marketplace adoption.

H3e: ICEC significantly influences business decision regarding e-marketplace adoption.

H4: Business attitude significantly influences business decision regarding e-marketplace adoption.

Hypotheses at the post-decision stage

H5: TAM constructs significantly influence adopting business decision continuity on e-marketplace adoption.

H5a: PU significantly influences adopting business decision continuity on e-marketplace adoption.

H5b: PEOU significantly influences adopting business decision continuity on e-marketplace adoption.

H5c: SN significantly influences adopting business decision continuity on e-marketplace adoption.

H5d: FC significantly influences adopting business decision continuity on e-marketplace adoption.

H5e: ICEC significantly influences adopting business decision continuity on e-marketplace adoption.

(9)

H6: TAM constructs significantly influence non-adopting business decision continuity on e-marketplace adoption.

H6a: PU significantly influences non-adopting business decision continuity on e-marketplace adoption.

H6b: PEOU significantly influences non-adopting business decision continuity on e-marketplace adoption.

H6c: SN significantly influences non-adopting business decision continuity on e-marketplace adoption.

H6d: FC significantly influences non-adopting business decision continuity on e-marketplace adoption.

H6e: ICEC significantly influences non-adopting business decision continuity on e-marketplace adoption.

H7: Business attitude significantly influences business decision continuity on e-marketplace adoption.

The five TAM constructs, PU, PEOU, firm characteristics (FC), SN and industry competitive environment characteristics (ICEC), will be formally defined in section 3.3. As can be seen from the above seven hypotheses, the hypotheses 1-2 aim to examine what influences firm attitude toward e-marketplace adoption, while the hypotheses 3-4 attempts to verify what influences firm decision on e-marketplace adoption. Hypotheses 5-7 set out to explore what influences firm decision continuity after adoption or non-adoption, which are determined by firms during the in-decision stage. Notably, the design of this study not only compensates for the observation of Karahanna et al. (1999) and Zain et al. (2005) that prevailing TAM-based research rarely distinguishes the correspondents into adopters and non-adopters, but also offers a dynamic view identifying the influences on business attitude toward e-marketplace adoption during the pre-decision stage, business decision to adopt e-marketplaces during the decision stage, and business adoption/ non-adoption continuance during the post-decision stage.

3.3 Justification of the TAM constructs

Before justifying the five TAM constructs in the research structure in details, we first define the five TAM constructs and briefly summarize how they were selected. Since the research structure and constructs are built on the theoretic basis of TAM and IDT, this research defines PEOU in terms of the degree to which the firm can effortless use e-marketplaces, PU in terms of benefits obtainable by the firm using e-marketplaces, FC as the degree to which a firm evaluates its current status, preparation, and

(10)

e-readiness as being able to assist in the adoption of an e-marketplace, ICEC as the degree to which a firm assesses the competitive environment in its industry as encouraging for the promotion of e-marketplace adoption, and SN as the degree to which the influence of the government, leading enterprises, and industry peer firms affect firm adoption of e-marketplaces.

Through careful examination of the literature on B2B e-marketplace adoption, we found that the e-marketplace characteristics play an important role influencing firm-level adoption of e-marketplaces. Regarding e-marketplace characteristics at the firm-level adoption context, adoption incentives such as cost reduction, revenue creation, transaction efficiency, increased competitiveness, expanded trading scope, and error reduction in trading processes (Bakos, 1991; Gottschalk & Abrahamsen, 2002; Holzmuller & Schluchter, 2002; White & Daniel, 2004; Yu, 2006) can be grouped into PU, while budget requirement, system standard, compatibility in both technological and non-technological issues, learning/training time, sunk cost, switching cost, and adoption barrier (Lee & Clark, 1996; Grewal et al., 2001; Stockdale & Standing, 2002; Ho et al., 2005; Zhu et al., 2006) can be grouped into PEOU.

Except for e-marketplace characteristics, the literature in Table 1 also reveals that the factors incurred from FC such as firm ability for the adoption of innovation IT/IS, business scale, degree of outsourcing, basic IT/IS infrastructure, and e-savviness of top management (Grewal et al., 2001; Stockdale & Standing, 2002; Ho et al., 2005; Yu, 2006), the factors incurred from ICEC such as market power, industry structure, business drivers, supplier enablement, buyer requirement, leading or forcing stakeholders (Lee & Clark, 1996; Stockdale & Standing, 2002; Holzmuller & Schluchter, 2002; Ganesh et al., 2004; Driedonks et al., 2005; Yu, 2006), and the factors incurred from SN such as positive experiences among adopters, knowledge exchange with opinion leaders and potential users, and government policy (Driedonks et al., 2005; Yu, 2006) all impact enterprise e-marketplace adoption.

In addition to the summarized high-level reasoning of how the five TAM constructs were derived, a more specific literature analysis of how related researching findings support the TAM constructs used in the proposed research model and hypotheses is needed. Accordingly, the following eight research studies are analyzed to map their outcomes to the eight TAM constructs.

(11)

the nature of organizational participation in an e-marketplace depends on a firm motivation and ability. Following empirical testing on 306 jewelry traders, Grewal et al. (2001) identified a positive relationship between the IT capability level and the likelihood of firm participation in e-marketplaces, with effort-based learning steering firms away from e-marketplace adoption. Accordingly, their study confirmed that PEOU is an antecedent of organizational decision in B2B e-marketplace participation, while FC such as organizational e-readiness significantly impacts the organizational decision in adopting/rejecting e-marketplaces.

Based on a survey of 65 Norwegian companies, Gottschalk and Abrahamsen (2002) concluded that reducing costs and gaining competitive advantages markedly encourage firm adoption of e-marketplaces. At the same year, Holzmuller and Schluchter (2002) surveyed 94 industry experts in Germany by a Delphi study and discovered that adoption of B2B e-marketplaces are motivated by increasing their competitiveness, for example by improving their business processes, and the top selection criteria for B2B e-marketplace relate to potential benefits. Accordingly, these two studies conducted in different countries both demonstrate that PU strongly influences enterprise e-marketplace adoption.

By interviewing managers of healthcare e-marketplaces in the UK as well as suppliers and buyers in those e-marketplaces, White and Daniel (2004) uncovered that enhancement of supplier-buyer relationships, reduction of errors and costs occurred in the transaction processes, and the time requirements to complete the transaction or respond the trader inquires are three critical influences on firm willingness to adopt e-marketplaces. The first influential factor reveals that industry counterparts can influence enterprise e-marketplace choice and adoption, an effect attributed to the influence of ICEC and SN. The second influential factor clearly belongs to PU, while the last one belongs to FC and ICEC.

Based on a case study of AuctionPlus conducted by the Australian Meat and Livestock Cooperation, Driedonks et al. (2005) applied economic and social perspectives to analyze the adoption of B2B e-marketplaces, and presented four observations. First, social networks are important in the adoption of B2B e-marketplaces. Second, communication channels are extremely important in the adoption of B2B e-marketplaces. Third, the existence of leading or powerful stakeholders in the industry considerably influences B2B e-marketplace adoption. Fourth, e-marketplace value is a function of user numbers.

(12)

The first observation implies that opportunities of business trades in certain industries rely on the exploitation of social networks such as common language and cultural, mutual understanding and trust, and so on (Driedonks et al., 2005). The second observation implies that positive experiences among early adopters and knowledge exchange with opinion leaders and counterparts extensively influence e-marketplace adoption. Meanwhile, the third observation implies that industries with leading or powerful stakeholders can facilitate e-marketplace adoption, and the fourth observation implies the effect of network externalities on e-marketplace adoption (Driedonks et al., 2005). Nevertheless, all observations appear to support the importance of SN in influencing business-level e-marketplace adoption, while the third observation also indicates the importance of ICEC (i.e., push or pressures from the adoption of major partners or competitors within the industry), while the fourth observation can be partly explained by PU.

Based on analysis of three published case studies and two in-depth case studies of government-supported e-marketplaces in Western Australia, Gengatharen and Standing (2005) found that the influences on the success or failure of government-supported regional e-marketplaces. Their work concluded that perceived benefits, relative advantages, top management commitment, firm internal readiness, firm size, government incentives, and normative pressures are key influences on the e-marketplace participation of small and medium enterprises. Clearly, the first two influences belong to PU, the following three influences support FC, and the final two influences are attributed to SN.

After comprehensively reviewing the literature on the adoption of electronic data interchange (EDI) systems, e-procurement systems, telecommunication and communication technologies, e-commerce, e-store, and other IT/IS, Yu (2006) categorized all possible influences into three constructs: pushes from outside the company, pulls from inside the company, and top-management e-savviness. After surveying 115 Taiwanese firms, Yu (2006) further identified that pushes from the partners or competitors within the industry, pushes from the government, extent of workflow computerization and standardization, time requirement, and degree of top management e-savviness significantly impact enterprise e-marketplace adoption. Obviously, the first two factors support SN and ICEC, the fourth factor can be attributed to FC and ICEC, and the remaining two factors can be attributed to FC.

(13)

Based on a survey of 329 managers and executives in Malaysian manufacturing firms, Zain et al. (2005) found that PEOU only affects firm attitude, while PU directly influences firm behavior, and a positive relationship exists between SN and firm behavior. However, in a study of 1562 employees of American firms that focused on enterprise ERP adoption, Amoako-Gyampah and Salam (2004) concluded that firm-level PEOU does not impact firm attitude and only PU influences firm attitude. Obviously, the findings of Amoako-Gyampah and Salam are not consistent with those of Zain et al. Since individual-level TAM has been extensively studied and ascertained by a considerable body of literature, firm-level TAM deserves more empirical reexaminations.

4. Questionnaire design, sampling, and analysis

This section contains three subsections. Notably, the instruments were prepared in Chinese but the items reproduced in this manuscript are English translations.

4.1 Construct Operationalizations

Given that Table 1 displayed that empirical studies regarding the adoption of B2B e-marketplaces are limited and e-marketplaces were evolved from EDI and developed based on e-procurement needs (Angeles 2000) and fully supported by IT, IS, and communication technologies (Guilherme & Aisbett 2003), to ensure the validity of the constructs used in this research, survey items were adapted not only from Table 1 but also from the pertinent literature, including EDI, e-procurement, telecommunication and communication, and IT/IS adoption (O' Callaghan et al., 1992; Premkumar et al., 1994; Iacovou et al., 1995; Premkumar and Roberts, 1995; Thong and Yap, 1995; Lai, 1998; Thong, 1999; Chau, 2001; Lucchetti and Sterlacchini, 2004) to operationalize the constructs of PU, PEOU, SN, FC, ICEC, and business-level attitude toward e-marketplace adoption as displayed in Fig. 3.

Accordingly, based on the works of Davis (1989), Davis et al.(1989), O' Callaghan et al. (1992), Premkumar et al. (1994), Premkumar and Roberts (1995), Iacovou et al., (1995), Thong and Yap (1995), Thong (1999), and Gottschalk and Abrahamsen (2002), PU was measured using eight items. The respondents were asked to indicate their level of agreement or disagreement with the following eight potential benefits of e-marketplace adoption:

(14)

1. Beneficial trading relationships with partners; 2. Enhanced collaboration with partners;

3. Increased competitive advantages; 4. Increased diversity of trading goods; 5. Increased source of buyers and sellers; 6. Increased speed of trade;

7. Increased opportunities to trade; and 8. Decreased trading costs.

PEOU was operationalized with four items derived from the works of Davis (1989), Davis et al. (1989), O' Callaghan et al. (1992), Premkumar et al. (1994), Thong and Yap (1995), and Thong (1999). The respondents were asked to indicate the extent to which they agreed with the statements related to e-marketplace adoption, which are as follows: 9. E-marketplace adoption requires a large capital investment in infrastructure

establishment.

10. E-marketplace adoption requires a large time investment in process restructuring. 11. E-marketplace adoption requires a large effort investment in training; and

12. E-marketplace adoption causes a large waste of investment in existing IS.

SN was assessed by eight items adapted from the works of O' Callaghan et al. (1992), Premkumar et al. (1994), Iacovou et al., (1995), Premkumar and Roberts (1999), Thong et al. (1995), Thong (1999), Lai (1998), White and Daniel (2004), and Driedonks et al. (2005). Respondents were asked to express their degree of agreement with the following statements using a seven-point Likert scale ranging from 1 (strongly disagree) to 7 (strongly agree):

13. A majority of leading enterprises within the supply chain use e-marketplaces. 14. A majority of trading parties within the supply chain use e-marketplaces. 15. A majority of peer competitors use e-marketplaces.

16. Leading firms in the industry recognize that an e-marketplace can enhance firm competitiveness.

17. Trading counterparts recognize that e-marketplaces can enhance firm competitiveness.

18. Peer competitors recognize that e-marketplaces can enhance firm competitiveness. 19. The government actively promotes e-marketplaces; and

20. E-marketplace adoption is supported by government grants.

(15)

Thong and Yap (1995), Thong (1999), Grewal et al. (2001), White and Daniel (2004), and Yu (2006), FC was assessed by 12 items, as listed below, on a seven-point Likert scale:

21. A majority of data communication tasks are processed via IS. 22. A majority of business reports are generated by IS.

23. A majority of problems are communicated via IS.

24. A majority of business processes are interconnected with IS. 25. All trade processes are clear and distinct.

26. All trade processes are documented.

27. All questions regarding trade processes can be answered using the documentation. 28. All trade processes are easy to computerize.

29. The timing for locating/attracting prospective traders is crucially important. 30. The timing for exchanging offerings with traders is crucially important. 31. The timing for instant communication with traders is crucially important, and 32. The timing for completing a transaction is crucially important.

Based on the studies of O' Callaghan et al. (1992), Grover and Goslar (1993), White and Daniel (2004), Ganesh et al. (2004), and Yu (2006), ICEC was operationalized by asking the respondents the following five questions:

33. Your firm is strongly influenced by the industry environment in terms of high-velocity competitive requirement.

34. Your firm is strongly influenced by the industry environment in terms of supplier enablement. 35. Your firm is strongly influenced by the industry environment in terms of buyer enablement. 36. Your firm is strongly influenced by the industry environment in terms of channel power. 37. Your firm is strongly influenced by the industry environment in terms of product power.

In contrast to using a single question to assess attitude as in most of the individual-level TAM literature, this work uses multiple items to operationalize business-level attitude due to business-level perception being a collective perception of the whole business’s decision-making members (Nelson & Quick, 2006). Moon et al. (2003) performed two group decision studies and empirically demonstrated that the decisions made by groups whose members with and without prior individual consideration of the problem exerted a different impact of the group decisions. The study of Moon et al. implies that technology adoption considerations of key stakeholders may markedly influence the business-level technology adoption. Consequently, referring to the works of Davis (1989), Davis et al. (1989), Moon et al. (2003), Zain et al. (2005), and Yu (2006), the business-level attitude toward

(16)

e-marketplace adoption was operationalized by asking respondents the following four questions:

38. The CEO strongly recognizes that e-marketplaces can enhance firm competitiveness. 39. The CEO has high awareness of e-marketplaces.

40. Senior management strongly recognizes that e-marketplaces can enhance firm competitiveness; and

41. Senior management has good awareness of e-marketplaces. 4.2 Sampling

Based on the above construct of operationalization drawn from the pertinent literature, the questionnaire comprises two sections. The first section contains 41 questions assessed using a Likert-type scale ranging from 1 (strongly disagree) to 7 (strongly agree), and collects the assessment of six constructs. The seven-point scale was chosen owing to it being more suitable for multi-variant analysis than smaller ranges, such as a five-point scale. The second section, containing nine questions listed in Appendix 1, gathers basic data on each respondent company and aims to determine whether or not the responding firm has joined an e-marketplace, whether or not the adopted firms plan to continue using, switch, or stop using e-marketplaces, and whether non-adopting firms plan to adopt e-marketplaces or not.

Before officially issuing questionnaires, a pretest was performed on scholars and participants of e-marketplaces to reword and refine the survey questions. Instead of mailing out the questionnaires, the pretest was conducted via face-to-face interviews to ensure that all questions and terms used could be clearly understood by respondents. Like past business-level TAM studies (Amoako-Gyampah and Salam, 2004; Zain et al., 2005) and the prevailing business-level academic surveys (Gatignon and Robertson, 1989; Chau and Tam, 2000; Fabiani et al., 2005; Zhu et al., 2006; Hadaya, 2006; Lancastre and Lages, 2006), this study used the key informant method. To ensure the representativeness of the respondents, clear and concise statements describing the purpose of the research were provided at the beginning of the questionnaire, and executives or managers who are responsible for firm e-marketplace adoption were invited to complete the questionnaire. The survey was issued to 1500 large-size firms randomly selected from the Top 5000 Company List published by China Credit Information Service LTD (http://www.credit.com.tw/newweb/DB/ index.htm).

(17)

Among 295 responses, 202 were considered valid, representing a valid response rate of 13.5%. Compared to survey return rates ranging from 11.5% to 16.5% for similar empirical studies on Taiwan industry during the recent years (Yu, 2006), a 13.5% valid response rate generated from a 19.7% total response rate was reasonable and compatible with recent surveys on Taiwanese firms. Table 2 briefly profiles the 202 firms as follows: 94 respondents had used at least one e-marketplace; these firms are highly profitable based on the ratio of revenue over capital exceeding 300 percent; the percentages of different industry type also suitably reflect the current industry distribution of the electronic, information, optics, machinery, metal, chemistry, and semiconductor industries in Taiwan. Moreover, the average annual membership fee of NT$ 42,194 (roughly US$1,310) is not high for large firms.

Insert Table 2 here.

4.3 Validity and reliability

Based on the research structure in Fig. 3, it appears reasonable to apply the structural equation model using a software such as LISREL for the hypothesis testing. However, there is a lack of sufficient theory-based literature investigating the antecedents of organizational participation in B2B e-marketplaces as listed in Table 1. Particularly, this investigation presents a multi-model research structure which is first drawn from TAM and IDT to investigate firm-level innovation technology adoption, namely e-marketplace. Therefore, this study approaches the data analysis through factor analysis to determine the construct validity and regression analysis for hypothesis testing.

The use of factor analysis is motivated by Yang and Yoo (2004), who claimed that attitude is important in TAM but has been ignored owing to the cognitive aspect that matters in TAM not being distinguished from the affection aspect in previous studies. The regression method is used because it can not only use a limited number of predictor variables to systematically clarify the tendency of the response variable, but can also quantify the relationship between dependent and independent variables and the explanatory power of the entire model (Neter et al., 1999). This may explain why Davis, who invented TAM in 1986, has always used the regression method to examine the hypotheses based on TAM, extended TAM, TAM 2, and unified TAM (Davis, 1989; Davis et al., 1989; Davis, 1993; Venkatesh & Davis, 1996 and 2000; Venkatesh et al., 2003).

(18)

Consequently, factor analysis using the principal component method with Varimax rotation was performed to verify whether or not the questionnaire items properly map the corresponding constructs. The criterion for each sorted question pertaining to each factor is that the Eigenvalue must exceed 1.0, and the corresponding intra-factor loading must exceed 0.6. After conducting the factor analysis using the SPSS software, questions 4 and 12 were removed because they failed to meet the above-mentioned criterion. Based on the results of factor analysis, there are two sub-constructs under PU and three sub-constructs under FC. Each sub-construct name is given by reflecting the context of the corresponding items as displayed in Table 3.

Insert Table 3 here.

As listed in the last column of Table 3, the computed Cronbach’s alpha coefficients for all dimensions exceed 0.867, indicating high content consistency between the questions relating to each of the constructs. Additionally, as displayed in Table 4, all of the inter-item correlation coefficients under each construct are significant (p < 0.001), revealing that the content has reliable, convergent, and discriminate properties (Davis, 1989; Adams et al., 1992). This study also examined the correlation coefficients among constructs and found positive correlations among the constructs. This finding demonstrated that firm attitude forms holistically rather than piecemeal, and indicates potential overlap among some of the constructs.

Insert Table 4 here.

5. Hypotheses test and data analysis

5.1 Hypothesis testing

Since regression is useful for forecasting the tendency of response variables in a systematic fashion with a limited number of explanatory variables (Neter et al., 1999), the linear regression method was used to verify hypotheses 1-2, while the logistic regression method was applied to examine hypotheses 3-7 due to their binary dependent variables with only two possible outcomes, namely yes and no. The computed results are summarized in Table 5.

(19)

In Model 1, the adjusted R2 demonstrates that 83.2 percent of firm attitude regarding e-marketplace adoption can be explained, while PEOU strongly influences PU in relation to business-level new technology adoption (p-value < 0.001). Additionally, the constructs of PU and SN considerably influence business attitude on e-marketplace adoption (p-value < 0.001), while PEOU and FC strongly influence business attitude toward e-marketplace adoption (p-value < 0.01), and ICEC does not statistically significantly affect business attitude toward e-marketplace adoption. Thus, hypotheses 1a-1d and 2 are accepted, and only hypothesis 1e is rejected.

One worthy-mentioned issue is the multicollinearity problem, which was not discussed in previous TAM studies using multiple regression method. Since PEOU significantly influencing PU, when both PEOU and PU are used to measure the relationship with Attitude in model 1, the problem of multicollineartiy indeed need to be checked up via the tolerance and the VIF values. In fact, multicollinearity is the property of every multiple regression model, but the multicollinearity problem occurs only when it significantly exceed a certain level. After resetting the SPSS software with the statistics option of multicollinearity diagnosis, the tolerance values are all greater 0.2 (0.321, 0.732, 0.338, 0.705, 0.624) and the VIF values are all smaller than 5 (3.114, 1.366, 2.961, 1.418, 1.603) for all the five independent variables, which, according to O'Brien (2007), are acceptable in practice.

Model 2 examines the relationship between the business decision regarding e-marketplace adoption and five constructs of PU, PEOU, SN, FC, and ICEC, as well as the relationship between the business decision and attitude regarding e-marketplace adoption. The empirical results indicate that the hypotheses 3a, 3c, and 4 are accepted, whereas hypotheses 3b and 3d-3e are rejected. That is, the positive relationship between the business attitude and decision is very significant, and among the five constructs only PU and SN significantly impact business decision on e-marketplace adoption.

Regarding Model 3, the positive relationship between business decision continuity and attitude is verified. That is, business adoption/non-adoption continuity is extremely significantly impacted by business attitude (p-value < 0.001). The decision continuity of the e-marketplace adopting businesses is very significantly influenced by PU (p-value < 0.01), and is also strongly influenced by FC and ICEC (p-value < 0.05). Meanwhile, for non-adopting businesses, the logistical regression results display that none of the five constructs considerably affect the decision continuity of the non-adopting businesses. In this case, hypotheses 5a, 5d-5e, and 7 are accepted, while hypotheses 5b-5c and 6a-6e

(20)

are rejected.

From the above analysis, at the pre-decision stage this study concluded that PU and SN are extremely important factors (p-values < 0.001), while PEOU and FC are very important factors (p-values < 0.01) influencing business attitudes. At the in-decision stage, this study verifies that only PU (p-values < 0.05) and SN (p-values < 0.01) significantly influence business attitudes, while business attitude is an extremely significant factor (p-values < 0.001) influencing business decision. At the post-decision stage, none among PU, PEOU, SN, FC, and ICEC significantly influence non-adopted business decision continuity about continuously not using e-marketplaces or planning to use e-marketplaces. However, PU very significantly influences (p-values < 0.01) and FC and ICEC significantly influence (p-values < 0.05) adopted business decision continuity regarding whether to continue using current e-marketplaces, based on which the following three conclusions can be reached. First, adopted businesses can maintain their original adoption decisions as long as they benefit from e-marketplaces adoption (i.e., the trading volume is growing due to using e-marketplaces, number of customers conducting transactions via e-marketplaces is growing, etc.). Second, the effect of PEOU is decreasing over time since businesses are used to using e-marketplaces. Third, the effect of SN is weakening over time owing to the increased first-hand experience of businesses with e-marketplaces.

Moreover, compared with the previous literature on B2B e-marketplace adoption, such as Yu (2006) which observed that external environmental pressures significantly influence e-marketplace adoption, this study discovered that ICEC did not significantly influence business attitude toward e-marketplace adoption during the pre-decision stage, and also not influenced business decisions regarding e-marketplace adoption during the in-decision stage. During the post-decision stage, ICEC did not significantly influence non-adopting business decision continuity, but ICEC significantly affected adopted business decision continuity (p-value < 0.05). This phenomenon can be attributed to ICEC only representing industry competitive environment in this study, because other external factors such the influence of government, the influence of leading or powerful stakeholders in the industry, and the like are covered by SN in this study. Besides, the survey units were small and medium companies in Yu (2006) while this work sampled large companies. Therefore, instead of saying that large firms are less impacted by the industry than small and medium firms, it may be more reasonable to say that small and medium businesses are more easily influenced by industry.

(21)

5.2 Adopted vs. non-adopted firm analysis

As Karahanna et al. (1999) noted, TAM studies typically examine the attitudes of all respondents regarding the IT adoption, but few studies have examined adopters (current customers) and non-adopters (called potential customers), respectively. Ignoring to examine the difference in views of both adopters and non-adopters toward innovation technology adoption may lead to certain valuable clues related to new technology adoption being lost. By using the t-test to examine the mean difference between adopters and non-adopters in terms of the constructs and sub-constructs, Table 6 shows the mean and standard deviation of these constructs for both adopters and non-adopters.

Insert Table 6 here.

For the macro level of the five constructs, Table 6 reveals that the scores on PEOU are especially low (average score less than 3.25) compared to those on other constructs (greater than 4.5 or even 5) for both adopted and not-adopted firms. This phenomenon may be attributed to the fact that although e-marketplaces are Internet-based open systems, their system interfaces and functionalities remain insufficiently friendly, which may partly explain why the adoption rate of e-marketplaces in Taiwan remains below expectations (Yu, 2006). From the micro level perspective, the figures regarding two sub-constructs of transaction functionality and competitive advantages under the PU in Table 6 demonstrate that transaction functionality does not impact business decisions on e-marketplace adoption, but competitive advantages are a crucial factor and lead to PU significantly influencing e-marketplace adoption.

6. Implications for theory and business

This study surveyed 202 large Taiwanese firms and thus derived four theoretical implications which paves the way for advancing current knowledge of the business-level new technology adoption. Meanwhile, some business implications are briefly described, which may be of interest to those who are interesting in effectively selling innovation IT/IS to enterprises.

6.1 Theoretical implications

(22)

explained by TAM or its variations is generally less than 40% (Hung et al., 2005), this empirical study have demonstrated that TAM possesses more powerful explanatory ability in B2B e-marketplace adoption (in which adjusted R2 is around 83%, as shown in Table 5). Accordingly, the first theoretical implication drawn from this empirical study is that TAM is not only useful in foreseeing individual-level new technology adoption but also works well in predicting business-level technology adoption (i.e., e-marketplace). This result may be attributed to collective business decisions involving more rationality and a longer decision time than single-person decisions. However, at the post-decision stage, the results of this study indicate that TAM cannot effectively explain business decision continuity following original adoption/non-adoption. Accordingly, to include other theories (i.e., expected-performance theory) into the research structure is required to understand what influences business-level decision continuity. Accordingly, the second theoretical implication is that TAM is only useful in explaining business attitude and decision regarding new technology adoption, but cannot predict business decision continuity after adoption/non-adoption.

The study of Moon et al. (2003) argued that the decisions made by groups whose members have and have not given prior individual consideration to the problem exert a different impact on group decisions. Similarly, many organizational behavior studies contending that organizational decision behavior has not only inherited the rational and irrational components of individual decisions but must also satisfy the concerns of multi-dimensional stakeholders (Nelson & Quick, 2006). Following the concept behind these studies, this research surveys 202 large Taiwanese firms regarding e-marketplace adoption. The empirical results demonstrate that enterprise beliefs significantly impact enterprise attitudes during the pre-adoption, and enterprise attitudes significantly impact enterprise decision and decision continuity during the in-decision and post-decision stages, respectively. Consequently, the third theoretical implication is that business-level attitude regarding innovation technology adoption during the pre-adoption can be effectively explained by business beliefs, and organization-level decision and decision continuity during the in-decision and post-decision are strongly influenced by business-level attitudes. Certainly, more empirical studies to examine the presented research model are necessary.

Looking at Tables 5 and 6, the figures have shown that SN significantly shapes business attitude before the decision and very strongly influences business decision during the decision itself. That is, potential adopters may acquire information from early adopters as well as being influenced by adopters with positive experience (Rogers, 2003)

(23)

or being forced by other important stakeholders in the industry (Driedonks et al., 2005). As argued by Frambach and Schillewaert (2002) and Moon et al. (2003), organizational-level attitude is the collective attitude of organization decision-makers, and organizational decision is made by a group of members who are influenced by their respective SN considerations of the problem just as for individual-level new technology adoption. Accordingly, this study confirmed that SN is not only a crucial factor influencing individual-level technology adoption, but also influences business-level technology adoption, which leads to the fourth theoretical implication.

6.2 Business implications

Recently, customer relationship management has received increasing attention, with a particular emphasis on the importance of understanding old customers. Although this work did not precisely identify the influences on the decision continuity of non-adopting firms at the post-decision stages, it identified the influences on the continued use of e-marketplaces by adopting firms. As shown in Table 5, three constructs of PU, FC, and ICEC significantly impact the initial decisions of adopting firms. Hence, for selling innovation IT/IS in the context of e-business, marketers can pay active attention to those firms with high usefulness expectations and low e-readiness and industry competitive environment. Meanwhile, some adopting firms with lower royalty and satisfaction can be identified, and then marketers should devise suitable service strategies to upgrade their satisfaction and royalty.

Furthermore, figures listed in Table 5 showed that attitude still significantly influences decisions of adopting firms to continue using current e-marketplaces or those of non-adopting firms to continue rejecting the use of e-marketplaces during the post-decision stage. Therefore, if an e-marketplace wishes to alter the original decisions of non-adopting firm, the marketers should first change firm attitudes. Similarly, if an e-marketplace wishes to retain current adopting firms, the original attitudes of those firms must be maintained and better enhanced to raise their royalty and usage. Firms will never change their original attitudes without some relevant motivation. Since determinants influencing business attitudes regarding e-marketplace adoption have been identified in this work, e-marketplace managers may feature their research and development, marketing, and service strategies based on those findings, and thus reverse the attitudes of non-adopting firms as well as strength the beliefs of adopting firms.

(24)

Table 5 demonstrates that PU, PEOU, SN, and FC significantly influence firm attitudes, while only PU and SN strongly influence business decisions. Hence, new technology marketers can prioritize their strategic focus subject to different stages such as selling innovation IT/IS to potential business-level customers at the pre-adoption stage, enhancing the willingness of existing users to continuously use their IT/IS at the post-adoption stage. This implies the theme of marketing programs should be adjusted as a subject of different stage to the users or prospective users, and the service programs must be effectively differentiated subject to different stages.

Moreover, combining Tables 5 and 6 together, this study found that SN exerts a much more powerful influence than the others. That is, the greater the effect of SN, the higher the likelihood of the firms adopting e-marketplaces. As a result, e-marketplace marketers may invite important stakeholders in the industry or leading/famous businesses to execute the testimonial marketing events to attract non-adopting firms to become adopters. Based on Table 6, the standardization and computerization are enablers via which FC significantly impacts business adoption of e-marketplaces. Consequently, marketers may select businesses with higher levels of workflow computerization and/or standardization as priority customers (the most prospective users).

Referring to previous TAM-based business-level studies (Zain et al., 2005; Amoako-Gyampah and Salam, 2004), Zain et al. (2005) concluded that PEOU affects only firm attitude and not PU, while Amoako-Gyampah and Salam (2004) discovered that PEOU significantly influences PU while only PU influences firm attitude. Compared to Zain et al. (2005) and Amoako-Gyampah and Salam (2004), this study found that PEOU significantly impacts PU, while both PU and PEOU significantly impacts company attitude. Regarding the relationship between PEOU and PU, this work supported Amoako-Gyampah and Salam (2004) instead of Zain et al. (2005), which may be attributed to the research object in Amoako-Gyampah and Salam (2004) is ERP system or inter-organizational IS, which are more complex than the desktop computer used as the research object in Zain et al. (2005). Accordingly, another business implication may result in as follows: the influence of PEOU on PU decreases with decreasing complexity of the new IT/IS-based products, the influence of PU is increasing with increasing complexity of the new IT/IS-based products. That is, the influence of PEOU on PU and business attitude and the influence of PU on business attitude are changeable and rely on the complexity of IT/IS itself.

(25)

Building the above discussion, when developing/launching an innovative IT/IS, the businesses should prioritize their marketing efforts on SN, focus their research and development efforts on PU and PEOU, and target their sales efforts on enterprises with higher levels of workflow computerization and standardization. Anyhow, the above discussion simply demonstrates some business implication drawn from this empirical study. To provide business with more useful clues, further elaborate research is needed. 7. Concluding Remarks

Like any study, this work naturally suffers certain limitations. First, due to the increasing number of higher education institutes over the last 15 years, from less than 30 initially to over 160 now, as well as to business and industry gradually moving to China (Tao et al., 2007), this study had practical difficulties in achieving a high response rate in the survey of Taiwanese firms. Future studies could overcome this problem by conducting qualitative case studies via in-depth face-to-face interviews to collect data, which could reexamine the business-level new IT/IS adoption model presented in this study. Second, this initial firm-level TAM study used regression method to analyze the collected data as Davis (1989) did, which is adequate for simple path models like the ones proposed in this study. However, for future insightful firm-level TAM studies, the Structured Equation Modelling (SEM)-base method is recommended for more sophisticated path models. Third, this was not a longitudinal study on examining the same sample from pre-adoption to the post-adoption stages. To verify the suitability of IDT for explaining business-level innovation diffusion, studies on a set of same samples from the pre-adoption, through the in-adoption, and finally the post-adoption stages are necessary. Finally, since the sample only obtained from Taiwanese enterprises and e-marketplace is an innovation IT/IS highlighting on the Internet context, caution is necessary in generalizing the methodology and findings to other technologies or other countries with different cultures or industrial structure.

However, by replacing the individual with the organization as an analysis unit in the measurement, this study confirms that TAM can effectively explain new technology adoption by enterprises. Compared to the vast individual-level TAM literature, business-level TAM literature is relatively scarce and has devoted insufficient attention to exploring the antecedents and consequences of business-level attitude and decision on new IT/IS adoption. Therefore, an important contribution of this study is to fill this gap, build links between TAM and business-level technology adoption, and pave the theoretical ground for advancing current understanding of business-level innovation technology adoption.

(26)

Besides, since e-marketplace is a web-based IT/IS, findings of this empirical study may provide interested parties with some clues about how to promote and sell web-based innovation technology to enterprise. Moreover, the adoption of e-marketplaces pertains to both IT and IS. The findings of this study thus can be generalized to the adoption of other business-level technology adoption. Finally, this investigation merely represents a preliminary work aimed to improve understanding of business-level innovation technology adoption from the perspective of behavioral theories. Further research is definitely required to verify and enhance the validity and generalization of the methodology used in this study.

References

1. Adams, D. A., Nelson, R. R., Todd, P. A., 1992. Perceived usefulness, ease of use, and usage of information. MIS Quarterly 16 (2), 227-247.

2. Ajzen, I., Fishbein, M., 1980. Understanding attitudes and predicting social behavior. Prentice-Hall, Englewood Cliffs, NJ.

3. Amoako-Gyampah, K., Salam, A. F., 2004. An extension of the technology acceptance model in an ERP implementation environment. Information & Management 41 (2), 731-745.

4. Angeles, R., 2000. Revisiting the role of Internet-EDI in the current electronic commerce scene. Logistics Information Management 13 (1), 45-50.

5. Bagozzi, R. P. 2007. The legacy of the technology acceptance model and a proposal for a paradigm shift. Journal of the Association for Information Systems 8(4), 244-254.

6. Bakos, J. Y., 1991. A strategic analysis of electronic marketplaces. MIS Quarterly 15 (3), 295-310.

7. Bakos, J. Y., 1997. Reducing buyer search costs: Implications for electronic marketplaces, Management Science 43 (12), 1676-1692.

8. Benbasat, I., Barki, H. 2007. Quo vadis, TAM? Journal of the Association for Information Systems 8(4), 211-218.

9. Benslimane, Y., Plaisent, M., Bernard, P., 2005. Investigating search costs and coordination costs in electronic markets: A transaction costs economics perspective. Electronic Markets 15 (3), 213-224.

10. Bhatacherjee, A., 2001. Understanding information system continuance: an expectation-confirmation model. MIS Quarterly 25(3), 351-370.

11. Chau, P. Y. K., 2001. Inhibitors to EDI adoption in small businesses: An empirical investigation. Journal of Electronic Commerce Research 2 (2), 78-88.

(27)

technology-push, technology-pull perspective. Information & Management 37 (2), 229-239.

13. Cooper, R. B., Zmud, R. W., 1990. Information technology implementation research: A technology diffusion approach. Management Science 36 (2), 123-139. 14. Davis, F. D., 1989. Perceived usefulness, perceived ease of use, and user of

acceptance of information technology. MIS Quarterly 13 (3), 319-340.

15. Davis, F. D., Bagozzi, R. P., Warshaw, P. R., 1989. User acceptance of computer technology: a comparison of two theoretical models. Management Science 35 (8), 982-1003.

16. Davis, F. D., 1993. User acceptance of information technology: System characteristics, user perceptions and behavioral impacts. International Journal of Man-Machine Studies 38 (3), 475-487.

17. Driedonks, C., Gregor, S., Wassenaar, A., Heck, E., 2005. Economic and social analysis of the adoption of B2B electronic marketplaces: A case study in the Australian beef industry. International Journal of Electronic Commerce 9 (3), 49-72.

18. Fabiani, S., Fabiano, S., Trento, S., 2005. ICT adoption in Italian manufacturing: Firm-level evidence. Industrial and Corporate Change 14 (2), 225-249.

19. Frambach, R. T., Schillewaert, N., 2002. Organizational innovation adoption: A multi-level framework of determinants and opportunities for future research. Journal of Business Research 55 (1), 163-176.

20. Ganesh, J., Madanmohan, T. R., Jose, P. D., Seshadri, S., 2004. Adaptive strategies of firms in high-velocity environments: The case of B2B electronic marketplaces. Journal of Global Information Management 12 (1), 41-59.

21. Gatignon, H., Robertson, T. S., 1989. Technology diffusion: An empirical test of competitive effects. Journal of Marketing 53 (1), 35-49.

22. Gengathren, D. E., Standing, C., 2005. A framework to assess the factors affecting success or failure of the implementation of government-supported regional e-marketplaces for SMEs. European Journal of Information Systems, 14, 417-433. 23. Gottschalk, P., Abrahamsen, A. F., 2002. Plans to utilize electronic marketplaces:

The case of B2B procurement markets in Norway. Industrial Management & Data Systems 102 (5), 325-331.

24. Grewal, R., Comer, J. M., Mehta, R., 2001. An investigation into the antecedents of organizational participation in business-to-business electronic markets, Journal of Marketing 65 (3), 17-33.

25. Grover, V., Goslar, M. D., 1993. The initiation, adoption, and implementation of telecommunications technologies in U.S. organization, Journal of Management

(28)

Information Systems 10 (1), 141-163.

26. Guilherme, D. P., Aisbett, J., 2003. The relationship between technology adoption and strategy in business-to-business markets: The case of e-commerce. Industrial Marketing Management 32 (4), 291-306.

27. Hadaya, P., 2006. Determinants of the future level of use of electronic marketplaces: The case of Canadian firms. Electronic Commerce Research 6 (2), 173-185.

28. Hartwick, J., Barki, H., 1994. Explaining the role of user participation in information system use. Management Science 40, 440-465.

29. Hernandez, B, Jimenez, J., Martin, M. J., 2008. Extending the technology acceptance model to include the IT decision-maker: A study of business management software, Technovation 28 (1), 112-121.

30. Ho, Y. C., Fu, H. P., Chien, P. H., 2005. Analysis of factors affecting adoption of the e-marketplace in Taiwan, Journal of the Chinese Institute of Industrial Engineers, 22 (5), 419-429.

31. Holzmuller, H. H., Schluchter, J., 2002. Delphi study about the future of B2B marketplaces in Germany, Electronic Commerce Research and Applications 1 (1), 2-19.

32. Hung, S. Y., Liang, T. P., Chang, C. M., 2005. A meta-analysis of empirical research using TAM. Journal of Information Management 12 (4), 211-233.

33. Iacovou, C. L., Benbasat, I., Dexter, A. S., 1995. Electronic data interchange and small organizations: Adoption and impact of technology. MIS Quarterly, 19 (4), 465-485.

34. Karahanna, E., Straub, D. W., Chervany, N. L., 1999. Information technology adoption across time: A cross-sectional comparison of pre-adoption and post-adoption beliefs. MIS Quarterly 23 (2), 183-213.

35. King, W. R., He, J., 2006. A meta-analysis of the technology acceptance model. Information & Management 43 (2), 740-755.

36. Lai, W. S., 1998. Studying the decision factors and execution status of enterprises in adopting e-store strategies, Master Thesis, Da-Ye University, Taiwan.

37. Lancastre, A., Lages, L. F., 2006. The relationship between buyer and a B2B e-marketplace: Cooperation determinants in an electronic market context. Industrial Marketing Management 35 (6), 774-789.

38. Lee, G. L., Clark, T. H., 1996. Market process reengineering through electronic market systems: Opportunities and challenges. Journal of Management Information Systems 13 (3), 113-136.

39. Lin, H.F., Lin, S.M., 2008. Determinants of e-business diffusion: A test of the technology diffusion perspective. Technovation 28 (1), 135-145.

(29)

40. Lucchetti, R., Sterlacchini, A., 2004. The adoption of ICT among SMEs: Evidence from an Italian Survey, Journal of Small Business 23 (2), 151-168.

41. Ma, Q., Liu, L. 2004. The technology acceptance model: a meta-analysis of empirical findings. Journal of Organization and End User Computing 16(1), 59-72. 42. Malone, T. W., Benjamin, R. I., Yates, J., 1987. Electronic markets and electronic

hierarchies: Effects of information technology on market structure and corporate strategies. Communications of ACM 30 (6), 484-497.

43. McFarlan, F. W., 1984. Information Technology Changes the Way You Compete. Harvard Business Review 62 (3), 98-103.

44. Moon, H., Conlon, D. E., Humphrey, S. E., Quingley, N., Devers, C. E., Nowakowski, J. M., 2003. Group decision process and incrementalism in organizational decision making, Organizational Behavior and Human Decision Processes 92, 67-79.

45. Nelson, D. L., Quick, J. C., 2006. Organizational Behavior: Foundations, Reality and Challenges. South-Western, Nashville, Tennessee.

46. Neter, J., Kutner, M. H., Nachtsheim, C. J., Wasserman, W., 1999. Applied Linear Regression Models (3rdedition), McGraw-Hill, Taipei, Taiwan.

47. O’Brian, R. M., 2007, A caution regarding rules of thumb for variance inflation factors, Quality and Quantity, 41(5), 673-690.

48. O’Callaghan, R. Kaufmann, R. J., Konsynski, B. R., 1992. Adoption correlates and share effects of electronic data interchange systems in marketing channel, Journal of Marketing 56, 45-56.

49. Premkumar, G., Ramamurthy, K., Nilakanta, S., 1994. Implementation of electronic data interchange: An innovation diffusion perspective, Journal of Management Information Systems 11 (2), 157-186.

50. Premkumar, G., Roberts, M., 1995. Adoption of computer aided software engineering (CASE) technology: An innovation adoption perspective, Data Base Advances 26 (2), 105-124.

51. Rogers, E. M., 2003. Diffusion of Innovations (5th edition), Free Press, New York. 52. Schepers, J., Wetzels, M. 2007. A meta-analysis of the technology acceptance

model: investigating subjective norm and moderation effects. Information Management 44, 90-103.

53. Stockdale, R., Standing, C., 2004. Benefits and barriers of electronic marketplace participation: An SME perspective, Journal of Enterprises Information Management 17 (4), 301-311.

54. Strader, T. J., Shaw, M. J., 1999. Consumer cost differences for traditional and Internet markets, Internet Research 9 (2), 82-92.

(30)

55. Tao, Y.-H., Chang, C.-R.. Chen, C.-P., 2007. Unmet adoption expectation as the key to e-marketplace failure: A case of Taiwan’s steel industry, Industrial Marketing Management 36, 1057-1067.

56. Thong, J. Y. L., Yap, C. S., 1995. CEO characteristics, organizational characteristics and information technology adoption in small business, Omega 23 (4), 429-442.

57. Thong, J. Y. L., 1999. An integrated model of information systems adoption in small business, Journal of Management Information Systems 15 (4), 187-214. 58. Teo, T.S.H., Ranganathan, C., Dhaliwal, J., 2006. Key dimensions of inhibitors for

the deployment of web-based business-to-business electronic commerce. IEEE Transactions on Engineering Management 53 (3), 395-411.

59. Venkatesh, V., Davis, F. D., 1996. A model of the antecedents of perceived ease of use: Development and test. Decision Sciences 27 (3), 451-481.

60. Venkatesh, V., Davis, F. D., 2000. A theoretical extension of the technology acceptance model: Four longitudinal field studies. Management Science 46 (2), 186-204.

61. 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. 62. Venkatesh, V., Davis, F. D., 2007. Dead or alive? The development, trajectory and

future of technology adoption research. Journal of the Association for Information Systems 8 (4), 267-286.

63. White, A., Daniel, E., 2004. Electronic marketplaces: An empirical study in the UK healthcare sector. International Journal of Electronic Business 2 (6), 603-623. 64. Yang, H. D., Yoo, Y., 2004. It’s all about attitude: Revisiting the technology

acceptance model. Decision Support Systems 28 (1), 19-31.

65. Yousafzai, S. Y., Gordon, R. F., Pallister, J. G., 2007. Technology acceptance: a meta-analysis of the TAM: Part 2. Journal of Modelling in Management 2(3), 281-304.

66. Yu, C. S., 2006. Exploring influences on Taiwanese SMEs e-marketplace adoption decision. Journal of Global Information Technology Management 9 (2), 5-21. 67. Zain, M., Rose, R. C., Abdullah, I., Masrom, M., 2005. The relationship between

information technology acceptance and organizational agility in Malaysia, Information & Management 42, 829-839.

68. Zhu, K., 2004. The complementarily of information technology infrastructure and e-commerce capability: a resource-based assessment of their business value. Journal of Management Information Systems 21 (1), 167-202.

數據

Table 1 Literature regarding the adoption of B2B e-marketplaces
Table 3 Factor analysis summary Construct Named sub-construct Q# Factor loading Eigenvalue Cumulatedvariance Cronbachα Transaction Functionality Q5Q6Q7 Q8 0.8820.9300.9030.842 3.282 46.89%PU Competitive Advantages Q1Q2 Q3 0.9360.9580.830 2.585 83.82% 0.867
Table 4: Inter-item correlation matrices Construct: PU Q1 1 Q2 Competitive Advantages 0.887 *** 1 Q3 0.692 *** 0.759 *** 1 Q5 0.258 ** 0.268 ** 0.427 ** 1 Q6 0.252 ** 0.252 ** 0.383 ** 0.821 *** 1 Q7 Transaction Functionality 0.158 * 0.159 * 0.333 ** 0.735
Table 5 Summary of regression test Results of Hypotheses 1-2 Model Dependent variables Independentvariables Standardized
+3

參考文獻

相關文件

Moreover, when compared with the battery charger with the traditional pulse-width-modulated one, the novel battery charger with zero-current switching converter indeed reduces

 develop a better understanding of the design and the features of the English Language curriculum with an emphasis on the senior secondary level;..  gain an insight into the

Consequently, Technology Education is characterized by learning activities which provide students with authentic experiences in various technological areas such as

3.1(c) again which leads to a contradiction to the level sets assumption. 3.10]) which indicates that the condition A on F may be the weakest assumption to guarantee bounded level

Compared with August 2006, the indices of Education, Miscellaneous goods &amp; services, and Housing &amp; fuels increased by 6.06%, 1.89% and 0.53% respectively, which were fuelled

The fuzzy model, adjustable with time, is first used to consider influence factors with different features such as macroeconomic factors, stock and futures technical indicators..

The New Knowledge-Infrastructure: The Role of Technology-Based Knowledge-Intensive Business Services in National Innovation Systems. Services and the Knowledge-Based

In terms of external cognitive factors, this research confirmed that assurance, apathy and price reasonability as part of the service quality dimension have influence on