understanding business-level innovation technology adoption · understanding business-level...
TRANSCRIPT
ARTICLE IN PRESS
0166-4972/$ - se
doi:10.1016/j.te
�Correspondfax: +8862 253
E-mail addr
Technovation 29 (2009) 92–109
www.elsevier.com/locate/technovation
Understanding business-level innovation technology adoption
Chian-Son Yua,�, Yu-Hui Taob
aInstitute of Information Technology and Management, Shih Chien University, #70 DaZhi Street, Taipei 10497, TaiwanbDepartment of Information Management, National University of Kaohsiung, Kaohsiung, Taiwan
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.
r 2008 Elsevier Ltd. All rights reserved.
Keywords: Technology acceptance model; Innovation diffusion theory; Electronic business; Electronic marketplace
1. Introduction
Accelerated growth of the Internet and electroniccommerce (e-commerce) in the recent decade has forcedbusinesses to encounter global competition and encouragedthem to establish a presence in global markets via theimplementation of new Internet-based information systemand technology (IT/IS). Given that the implementation ofnew Internet-based IT/IS is a continued adoption processfor transforming a business toward electronic business(e-business), establishing electronic links with its suppliersand buyers, and executing electronic transactions alongvalue-chain activities, the business attitudes regardingthe adoption of innovation IT/IS have been recognizedas a critical factor for executing e-business strategy.However, e-business is different from previous traditionaltechnological innovation (Lin and Lin, 2008). In contrast,e-business represents a new innovative approach to incor-
e front matter r 2008 Elsevier Ltd. All rights reserved.
chnovation.2008.07.007
ing author. Tel.: +886 2 25381111x1700, 1701, 8921;
33143.
ess: [email protected] (C.-S. Yu).
porate core business processes/functions with Internet-basedIT/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 heavilybased on behavioral theories such as technology accep-tance model (TAM), theory of planned behavior (TPB),and innovation diffusion theory (IDT) (Hernandez et al.,2008). Literature on business-level technology adoption isscarce compared to general literature on examiningindividual-level technology adoption, and in particularcontains few studies adopting the TAM standpoint.Enterprises allocate significant portions of their budgeteach year to procuring new IT/IS, and this trend hasbecome more obvious following the advance of IT/IS andthe diffusion and development of the e-life, e-society, ande-business. Hence, understanding business-level innovativetechnology adoption is just as important as understandingindividual-level new technology adoption.This study chooses electronic marketplace (e-market-
place) as the study object, because the e-marketplace is anInternet-based IT/IS innovation and application. Althoughrapid growth of e-marketplaces appears inevitable, asurvey undertaken in early 2004 (Yu, 2006) indicated that
ARTICLE IN PRESS
PU
PEOUAttitude to
useActual usebehavior
SN
Fig. 1. A business-level technology acceptance model.
C.-S. Yu, Y.-H. Tao / Technovation 29 (2009) 92–109 93
the adoption rate of e-marketplaces by Taiwanese enter-prises was approximately 23.48%, which was well belowexpectations. Therefore, the reasons why certain enter-prises wish to use the e-marketplaces, while others do not,are quite interesting to be investigated. Compared to thevast existing literature on e-marketplaces, relatively fewworks have studied business to business (B2B) e-market-places adoption from behavioral theories, and most suchstudies are conducted from the economic viewpoint(Bakos, 1991, 1997; Strader and Shaw, 1999; Benslimaneet al., 2005; Zhu et al., 2006) which may not fully explainthe B2B e-marketplace adoption (Driedonks et al., 2005).
Motivated by the above discussion, this study focuses onthree objectives. First, this work examines how business-level attitudes influence innovative Internet-based IT/ISadoption. Second, in contrast to the previous TAM-basedliterature, which generally takes individual-level users asthe survey unit, this study takes collective organizations asthe analysis unit to examine whether TAM remains valid atthe business-level technology adoption. Third, rather thanusing an economic perspective, and only providing a staticview in examining the influences on e-marketplace adop-tion by enterprises, this paper integrates TAM and IDT toform a multi-model research structure that reveals adynamic picture of a firm’s attitudes before new technologyadoption, decisions to adopt the new technology, anddecisions to continue using or rejecting it.
2. Theoretical framework
Compared to the large body of individual-level TAMliterature, business-level TAM literature is relatively rare,though studies using TAM to examine organizational-leveltechnology adoption are not entirely novel (Amoako-Gyampah and Salam, 2004; Zain et al., 2005). However,such researches have failed to clarify the relationshipbetween business-level attitude and behavior on innovativetechnology adoption. This implies that the underlyingtechnology adoption at the firm level has not beendiscussed and ascertained in sufficient detail.
2.1. TAM
TAM, proposed by Davis in 1986 (Davis, 1989), is usedto effectively forecast individual computer acceptancebehavior, and was adapted from theory of reasoned action(TRA) developed by Ajzen and Fishbein in 1975 (Ajzenand Fishbein, 1980). In TAM, the actual behavior (AB) ofan individual to adopt a technology-based product can bepredicted by the perceived usefulness (PU) and perceivedease-of-use (PEOU) of that individual as expressed by theregression model AB ¼ b1 PU+b2 PEOU+e, where PU isdefined as the subjective assessment of a user or prospectiveuser that using the product will provide benefits related tojob performance, and PEOU is the degree to which anindividual 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 ofindividuals, the usefulness of business-level TAM can bedefined as the number of benefits obtainable by thecompany using the new technology, which is subjectivelyevaluated by key decision makers in firms. Likewise, ease-of-use can be defined as the degree to which business caneffortlessly use the new technology. Effort in this contextcan refer to monetary investment, employee trainingtime, technology switching barriers, maintenance costs,and so on.Over the past two decades, enormous studies have used
TAM or related extensions to provide empirical evidenceon the relationships among PU, PEOU, and AB, or tovalidate and enhance the reliability and robustness of theTAM questionnaire instrument. Notably, although Daviset al. (1989) argued that the subjective norm (SN) did notsignificantly influence usage intention, and thus omitted SNin their original TAM, Davis modified this approach(Venkatesh and Davis, 2000) and concluded that SNconsiderably influences the attitude toward IT productadoption, based on numerous empirical studies demon-strating this (Hartwick and Barki, 1994; Karahanna et al.,1999). From an organizational behavior perspectives,many studies found that organizational decision behaviornot only inherits the rational and irrational components ofindividual decisions but is also a collective perceptionreflecting the concerns of multi-dimensional stakeholders(Frambach and Schillewaert, 2002; Nelson and Quick,2006).Building on the above discussion, the new technology
adoption behavior demonstrated by the whole businessmight resemble that demonstrated by a single individual.Accordingly, generalized business-level technology adop-tion attitude and behavior maybe can also be effectivelyexplained by TAM, as shown in Fig. 1. Since this studyexplores the influences of firm attitude, decision, andcontinuance on e-marketplace adoption/non-adoption, theterm ‘‘consumer’’ as used in the remainder of this papermay refer to a business, firm, or organization.
2.2. IDT
IDT, pioneered by Rogers in 1962 (Rogers, 2003), is usedas a process-oriented perspective to explain how aninnovation can be accepted and disseminated amongconsumers. IDT contends that the adoption or rejectionof an innovation begins with consumer awareness of that
ARTICLE IN PRESS
ConfirmationDecision Implementation Knowledge Persuasion
Characteristics of
decision-making
unit
Perceived
characteristics of
the innovation
Adoption
or
rejection
Continued adoption
Late adoption
Discontinuance
Continued rejection
Pre-decision In-decision Post-decision
Fig. 2. Diffusion of innovation (Rogers, 2003).
Table 1
Literature regarding the adoption of B2B e-marketplaces
Sources Theory Study type
Bakos (1991) Economic theory Non-empirical
Lee and Clark
(1996)
None Non-empirical
Grewal et al.
(2001)
Organizational
motivation and
ability
Empirical (collecting data
from 306 jewelry traders)
Gottschalk and
Abrahamsen
(2002)
None Empirical (collecting data
from 65 companies in
Norway)
Stockdale and
Standing (2004)
None Non-empirical
Holzmuller and
Schluchter (2002)
None Empirical (collecting data
from 94 experts in Germany)
White and Daniel
(2004)
None Empirical (interviewing
managers, suppliers, and
buyers of healthcare e-
marketplaces in UK)
Stockdale and
Standing (2004)
None Non-empirical
Driedonks et al.
(2005)
Market theory
and DOI
Case study
Ho et al. (2005) None Non-empirical
Gengathren and
Standing (2005
Institutional
theory
Case study
Yu (2006) None Empirical (collecting data
from 115 companies in
Taiwan)
C.-S. Yu, Y.-H. Tao / Technovation 29 (2009) 92–10994
innovation, while diffusion is a process in which informa-tion regarding an innovation is conveyed via certainchannels over time among consumers. Innovation isdefined as an idea, practice, product, or object thatconsumers perceive as new. Despite receiving numerouspositive and negative criticisms corresponding to differentcognitive styles, IDT has been applied to over thousands ofempirical studies since 1962 (Rogers, 2003) includingstudies of business-level innovation technology adoption(Gatignon and Robertson, 1989; Cooper and Zmud, 1990;Chau and Tam, 2000; Frambach and Schillewaert, 2002).Because many innovations require a lengthy period beforebeing popular, which frequently lasts for many years, IDTis quite useful in identifying ways to accelerate innovationdiffusion from the time they become available to the timethey are widely adopted (Rogers, 2003).
IDT maintains that, during the pre-decision stage,consumers actively seek and/or passively receive informa-tion regarding innovations and shape their favorable orunfavorable beliefs regarding the innovations. The in-decision stage takes place when consumers engage inactivities that lead them to make a choice between adoptingand rejecting an innovation. Meanwhile, the post-decisionstage occurs immediately after consumers begin using orreject using an innovation. During the post-decision stage,consumers seek reinforcement for their previous decision,and may reverse the decision if exposed to relateddissonance messages. Restated, non-adopters may eithercontinuously reject the use of the innovation or choose toadopt it. Non-adopters adopt an innovation if motivatedto do so after securing further information or evidence thatinfluences their original beliefs of not adopting theinnovation. Conversely, adopters may continuously usethe innovation or alternatively may stop using it owing todissatisfaction with its performance. Fig. 2 shows the IDTprocess of Rogers.
3. Research structure
E-marketplace development has only just begun itssteady growth during the recent years, and represents aninnovation Internet-based IT, IS, or business model.However, the initial idea of establishing a cybernetic
buying and selling platform was originally presented byseveral authors during the period long before the inceptionof the Internet or e-commerce (McFarlan, 1984; Maloneet al., 1987; Bakos, 1991). Despite the existence of a largebody of e-marketplace studies and literature regarding B2Be-marketplace adoption dating back to the early 1990s(Bakos, 1991; Lee and Clark, 1996), literature directlyrelated to firm-level e-marketplaces is relatively scarce andis briefly summarized in Table 1.Instead of describing the development of the research
model, the paper presents the formulated research modelbased on the research objectives set up in the ‘‘Introduc-tion’’ section, and the principles established next in thissection. Also, the research hypotheses are derived with theresearch model concurrently, followed by the justification
ARTICLE IN PRESS
Pre-decision In-decision Post-decision
C.-S. Yu, Y.-H. Tao / Technovation 29 (2009) 92–109 95
of the determinants proposed in the research model andhypotheses.
H6
H5
Model 3 testing business decision continuity at post-decision stage
H1
Model 1 testing business attitude at pre-decision stage
H7
H4
H3
H2
FC
ICEC
PU
PEOU
SNAttitude
ICEC
PU
PEOU
SN
FC
ICEC
PU
PEOU
SNDecision
Continuity
FC
Model 2 testing business decision at in-decision stage
Adoptiondecision
Attitude
Attitude
Fig. 3. The proposed research structure.
3.1. Principles of structure formulation
The beauty of Davis’ TAM is simple and effective byemploying two simple constructs, PU and PEOU, toreplace many of TRA constructs while being ableto explain the adoption behavior from the perspective ofeither the critical reviews (Bagozzi, 2007; Benbasat andBarki, 2007) or meta-analyses (King and He 2006;Yousafzai et al., 2007). Because of this merit, there wasplenty of room for many extended TAM studies since itsinception, which according to Bagozzi (2007) had attractedan incredible number of over 700 citations. Similarly, as aninitial study on firm-level TAM adoption behavior, theresearch model should not be too complicated, which notonly provides a solid ground for extending studies but alsoprevent this model from being trivial in its scope asconcluded by some of the critical reviews about recentTAM-related studies (Venkatesh and Davis, 2007; Bagozzi,2007; Benbasat and Barki, 2007). Accordingly, the firstprinciple is to make the fundamental research structure assimple as possible for providing a solid research base forfuture firm-level TAM studies.
Compared to extant e-marketplace literature relyingheavily on an economic perspective and providingonly a static view in examining the influences one-marketplace adoption by enterprises, this paper attemptsto integrate TAM and IDT to form a multi-modelresearch structure that reveals a dynamic picture of afirm’s attitudes before new technology adoption, decisionsto adopt the new technology, and decisions to continueusing or rejecting it. Therefore, the second principle is toobserve firm-level adoption at different decision pointsderived from the TAM and IDT from a process-orientedperspective.
3.2. Research structure and hypotheses
First of all, grounded in TAM and IDT and based on thesecond principle of structure formulation, there are threemodels corresponding to the decision points, including pre-decision, in-decision, and post-decision, as depicted inFig. 3. Compared to Davis’s TAM model (1989), thismodel further considered behavior at the adoption spanand after the adoption, which provides a more precisedistinguishability of the critical factors that influence thefirm behavior from a process perspective. Accordingly, thehypotheses are divided into three sets at different decisionpoints, and posited from TAM and the literature analysisof potential factors influencing the decision of B2Be-marketplace adoption, which is discussed in Section 3.3.A brief rationalization behind the structure formulation isdescribed in this subsection, and some model componentsare justified by the literature in the next subsection.
As depicted in Fig. 3, each model has a differentdependent variable of concerns during the process, as alsoimplied by the first principle of structure formulation. InDavis’s TAM model (1989), the antecedent factorsinfluences three different-stage dependent variables, i.e.,attitude, intention to adopt, and AB, respectively. Accord-ingly, attitude is an observation before the adoptiondecision as seen in model 1, while the adoption decisionitself is also influenced by the attitude as seen in model 2.Moreover, the behavior after decision may be differentthan the previous two, as seen in model 3, which accordingto Venkatesh and Davis (2007) is also an emergent researchfocus on continuous usage of IT/IS introduced byBhatacherjee (2001). These different dependent variablescan also be clearly seen in the three sets of hypotheses.Notably, in order to observe the different dependent
variables, all three models employ the same set of TAMconstructs, which will be justified in the next subsection.Nevertheless, by applying the first principle of structureformulation, this study does not intend to exhaustivelycover all the constructs in previous TAM studies. Mean-while, this study does not investigate their mutual relation-ships since it largely complicated the models. However,
ARTICLE IN PRESSC.-S. Yu, Y.-H. Tao / Technovation 29 (2009) 92–10996
PEOU–PU relationship in pre-decision is included since ithas been examined by almost all individual-level TAMmeta-analyses (Ma and Liu, 2004; King and He, 2006;Yousafzai et al., 2007; Schepers and Wetzels, 2007). In fact,the idea of using TAM to examine business-level technol-ogy adoption is not entirely new (Amoako-Gyampah andSalam, 2004; Zain et al., 2005). Based on a survey of 329managers and executives in Malaysian manufacturingfirms, Zain et al. (2005) found that PEOU does not affectfirm-level PU, while in a study of 1562 employees ofAmerican 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 andSalam is not consistent, which is worth an investigation inthis firm-level TAM study.
Finally, the post-decision model distinguishes adoptersfrom non-adopters since it has to occur before the post-decision point of continue usage. That is why there arethree hypotheses in model 3 compared to the similar model2. It makes sense to understand those who did not adopt atthe first place but changed their mind afterward sincevaluable factors are hidden in this change behavior.
Built on the above, the hypotheses corresponding tothese three models are posited as follows:
Hypotheses at the pre-decision stage
H1: TAM constructs significantly influence businessattitude to e-marketplace adoption.
H1a: PU significantly influences business attitudeto e-marketplace adoption.H1b: PEOU significantly influences businessattitude to e-marketplace adoption.H1c: SN significantly influences business attitudeto e-marketplace adoption.H1d: FC significantly influences business attitudeto e-marketplace adoption.H1e: ICEC significantly influences business atti-tude to e-marketplace adoption.
H2: Business PEOU significantly influences its PU.
Hypotheses at the in-decision stage
H3: TAM constructs significantly influence businessdecision regarding e-marketplace adoption.
H3a: PU significantly influences business decisionregarding e-marketplace adoption.H3b: PEOU significantly influences business deci-sion regarding e-marketplace adoption.H3c: SN significantly influences business decisionregarding e-marketplace adoption.H3d: FC significantly influences business decisionregarding e-marketplace adoption.H3e: ICEC significantly influences business deci-sion regarding e-marketplace adoption.
H4: Business attitude significantly influences businessdecision regarding e-marketplace adoption.
Hypotheses at the post-decision stage
H5: TAM constructs significantly influence adoptingbusiness decision continuity on e-marketplace adop-tion.
H5a: PU significantly influences adopting businessdecision continuity on e-marketplace adoption.H5b: PEOU significantly influences adoptingbusiness decision continuity on e-marketplaceadoption.H5c: SN significantly influences adopting businessdecision continuity on e-marketplace adoption.H5d: FC significantly influences adopting businessdecision continuity on e-marketplace adoption.H5e: ICEC significantly influences adopting busi-ness decision continuity on e-marketplace adop-tion.
H6: TAM constructs significantly influence non-adopting business decision continuity on e-market-place adoption.
H6a: PU significantly influences non-adoptingbusiness decision continuity on e-marketplaceadoption.H6b: PEOU significantly influences non-adoptingbusiness decision continuity on e-marketplaceadoption.H6c: SN significantly influences non-adoptingbusiness decision continuity on e-marketplaceadoption.H6d: FC significantly influences non-adoptingbusiness decision continuity on e-marketplaceadoption.H6e: ICEC significantly influences non-adoptingbusiness decision continuity on e-marketplaceadoption.
H7: Business attitude significantly influences businessdecision continuity on e-marketplace adoption.
The five TAM constructs: PU, PEOU, firm character-istics (FC), SN, and industry competitive environmentcharacteristics (ICEC), will be formally defined in Section3.3. As can be seen from the above seven hypotheses, thehypotheses 1 and 2 aim to examine what influences firmattitude toward e-marketplace adoption, while the hypoth-eses 3 and 4 attempt to verify what influences firm decisionon e-marketplace adoption. Hypotheses 5–7 set out toexplore what influences firm decision continuity afteradoption or non-adoption, which are determined by firmsduring the in-decision stage. Notably, the design of thisstudy not only compensates for the observation ofKarahanna et al. (1999) and Zain et al. (2005) thatprevailing TAM-based research rarely distinguishes thecorrespondents into adopters and non-adopters, but alsooffers a dynamic view identifying the influences on businessattitude toward e-marketplace adoption during the pre-decision stage, business decision to adopt e-marketplacesduring the decision stage, and business adoption/non-adoption continuance during the post-decision stage.
ARTICLE IN PRESSC.-S. Yu, Y.-H. Tao / Technovation 29 (2009) 92–109 97
3.3. Justification of the TAM constructs
Before justifying the five TAM constructs in the researchstructure in details, we first define the five TAM constructsand briefly summarize how they were selected. Since theresearch structure and constructs are built on the theoreticbasis of TAM and IDT, this research defines PEOU interms of the degree to which the firm can effortlessly usee-marketplaces, PU in terms of benefits obtainable by thefirm using e-marketplaces, FC as the degree to which a firmevaluates its current status, preparation, and e-readiness inbeing able to assist in the adoption of an e-marketplace,ICEC as the degree to which a firm assesses the competitiveenvironment in its industry as encouraging for thepromotion of e-marketplace adoption, and SN as thedegree to which the influence of the government, leadingenterprises, and industry peer firms affect firm adoption ofe-marketplaces.
Through careful examination of the literature on B2Be-marketplace adoption, we found that the e-marketplacecharacteristics play an important role influencing firm-leveladoption of e-marketplaces. Regarding e-marketplacecharacteristics at the firm-level adoption context, adoptionincentives such as cost reduction, revenue creation,transaction efficiency, increased competitiveness, expandedtrading scope, and error reduction in trading processes(Bakos, 1991; Gottschalk and Abrahamsen, 2002;Holzmuller and Schluchter, 2002; White and Daniel,2004; Yu, 2006) can be grouped into PU, while budgetrequirement, system standard, compatibility in bothtechnological and non-technological issues, learning/train-ing time, sunk cost, switching cost, and adoption barrier(Lee and Clark, 1996; Grewal et al., 2001; Stockdale andStanding, 2004; Ho et al., 2005; Zhu et al., 2006) can begrouped into PEOU.
Except for e-marketplace characteristics, the literaturein Table 1 also reveals that the factors incurred from FCsuch as ability of a firm for the adoption of innovationIT/IS, business scale, degree of outsourcing, basicIT/IS infrastructure, and e-savviness of top management(Grewal et al., 2001; Stockdale and Standing, 2004; Hoet al., 2005; Yu, 2006), the factors incurred from ICECsuch as market power, industry structure, business drivers,supplier enablement, buyer requirement, leading or forcingstakeholders (Lee and Clark, 1996; Stockdale and Stand-ing, 2004; Holzmuller and Schluchter, 2002; Ganesh et al.,2004; Driedonks et al., 2005; Yu, 2006), and the factorsincurred from SN such as positive experiences amongadopters, knowledge exchange with opinion leaders andpotential users, and government policy (Driedonks et al.,2005; Yu, 2006) all impact enterprise e-marketplaceadoption.
In addition to the summarized high-level reasoning ofhow the five TAM constructs were derived, a more specificliterature analysis of how related researching findingssupport the TAM constructs used in the proposed researchmodel and hypotheses is needed. Accordingly, the follow-
ing eight research studies are analyzed to map theiroutcomes to the five TAM constructs.Relying on the motivation–ability framework, Grewal
et al. (2001) theorized that the nature of organizationalparticipation in an e-marketplace depends on a firmmotivation and ability. Following empirical testing on306 jewelry traders, Grewal et al. (2001) identified apositive relationship between the IT capability level and thelikelihood of firm participation in e-marketplaces, witheffort-based learning steering firms away from e-market-place adoption. Accordingly, their study confirmed thatPEOU is an antecedent of organizational decision in B2Be-marketplace participation, while FC such as organiza-tional e-readiness significantly impacts the organizationaldecision in adopting/rejecting e-marketplaces.Based on a survey of 65 Norwegian companies,
Gottschalk and Abrahamsen (2002) concluded that redu-cing costs and gaining competitive advantages markedlyencourage firm adoption of e-marketplaces. At the sameyear, Holzmuller and Schluchter (2002) surveyed 94industry experts in Germany by a Delphi study anddiscovered that adoption of B2B e-marketplaces ismotivated by increasing their competitiveness, for exampleby improving their business processes, and the top selectioncriteria for B2B e-marketplace relate to potential benefits.Accordingly, these two studies conducted in differentcountries both demonstrate that PU strongly influencesenterprise e-marketplace adoption.By interviewing managers of healthcare e-marketplaces
in the UK as well as suppliers and buyers in thosee-marketplaces, White and Daniel (2004) uncovered thatenhancement of supplier–buyer relationships, reduction oferrors and costs occurred in the transaction processes, andthe time requirements to complete the transaction orrespond to the trader inquires are three critical influenceson firm willingness to adopt e-marketplaces. The firstinfluential factor reveals that industry counterparts caninfluence enterprise e-marketplace choice and adoption, aneffect attributed to the influence of ICEC and SN. Thesecond influential factor clearly belongs to PU, while thelast one belongs to FC and ICEC.Based on a case study of AuctionPlus conducted by the
Australian Meat and Livestock Cooperation, Driedonkset al. (2005) applied economic and social perspectives toanalyze the adoption of B2B e-marketplaces, and presentedfour observations. First, social networks are important inthe adoption of B2B e-marketplaces. Second, communica-tion channels are extremely important in the adoption ofB2B e-marketplaces. Third, the existence of leading orpowerful stakeholders in the industry considerably influ-ences B2B e-marketplace adoption. Fourth, e-marketplacevalue is a function of user numbers.The first observation implies that opportunities of
business trades in certain industries rely on the exploitationof social networks such as common language and cultural,mutual understanding and trust, and so on (Driedonkset al., 2005). The second observation implies that positive
ARTICLE IN PRESSC.-S. Yu, Y.-H. Tao / Technovation 29 (2009) 92–10998
experiences among early adopters and knowledge exchangewith opinion leaders and counterparts extensively influencee-marketplace adoption. Meanwhile, the third observationimplies that industries with leading or powerful stake-holders can facilitate e-marketplace adoption, and thefourth observation implies the effect of network external-ities on e-marketplace adoption (Driedonks et al., 2005).Nevertheless, all observations appear to support theimportance of SN in influencing business-level e-market-place adoption, while the third observation also indicatesthe importance of ICEC (i.e., push or pressures from theadoption of major partners or competitors within theindustry), while the fourth observation can be partlyexplained by PU.
Based on analysis of three published case studies and twoin-depth case studies of government-supported e-market-places in Western Australia, Gengathren and Standing(2005) found the influences on the success or failure ofgovernment-supported regional e-marketplaces. Theirwork concluded that perceived benefits, relative advan-tages, top-management commitment, firm internal readi-ness, firm size, government incentives, and normativepressures are key influences on the e-marketplace partici-pation of small and medium enterprises. Clearly, the firsttwo influences belong to PU, the following three influencessupport FC, and the final two influences are attributedto SN.
After comprehensively reviewing the literature on theadoption of electronic data interchange (EDI) systems,e-procurement systems, telecommunication and commu-nication technologies, e-commerce, e-store, and otherIT/IS, Yu (2006) categorized all possible influences intothree constructs: pushes from outside the company, pullsfrom inside the company, and top-management e-savviness.After surveying 115 Taiwanese firms, Yu (2006) furtheridentified that pushes from the partners or competitorswithin the industry, pushes from the government, extent ofworkflow computerization and standardization, time re-quirement, and degree of top-management e-savvinesssignificantly impact enterprise e-marketplace adoption.Obviously, the first two factors support SN and ICEC, thefourth factor can be attributed to FC and ICEC, and theremaining two factors can be attributed to FC.
Based on a survey of 329 managers and executives inMalaysian manufacturing firms, Zain et al. (2005) foundthat PEOU affects only firm attitude, while PU directlyinfluences firm behavior, and a positive relationship existsbetween SN and firm behavior. However, in a study of1562 employees of American firms that focused onenterprise ERP adoption, Amoako-Gyampah and Salam(2004) concluded that firm-level PEOU does not impactfirm attitude and only PU influences firm attitude.Obviously, the findings of Amoako-Gyampah and Salamare not consistent with those of Zain et al. Since individual-level TAM has been extensively studied and ascertained bya considerable body of literature, firm-level TAM deservesmore empirical reexaminations.
4. Questionnaire design, sampling, and analysis
This section contains three subsections. Notably, theinstruments were prepared in Chinese but the itemsreproduced in this manuscript are English translations.
4.1. Construct operationalizations
Given that Table 1 displayed that empirical studies regardingthe adoption of B2B e-marketplaces are limited and e-marketplaces were evolved from EDI and developed basedon e-procurement needs (Angeles, 2000) and fully supported byIT, IS, and communication technologies (Guilherme andAisbett, 2003), to ensure the validity of the constructs used inthis research, survey items were adapted not only from Table 1but also from the pertinent literature, including EDI,e-procurement, telecommunication and communication, andIT/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 theconstructs of PU, PEOU, SN, FC, ICEC, and business-levelattitude 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), andGottschalk and Abrahamsen (2002), PU was measuredusing eight items. The respondents were asked to indicatetheir level of agreement or disagreement with the followingeight potential benefits of e-marketplace adoption:
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 derivedfrom 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). Therespondents were asked to indicate the extent to whichthey agreed with the statements related to e-market-place adoption, which are as follows:
9.
e-marketplace adoption requires a large capitalinvestment in infrastructure establishment;10.
e-marketplace adoption requires a large time invest-ment in process restructuring;11.
e-marketplace adoption requires a large effort invest-ment in training; and12.
e-marketplace adoption causes a large waste ofinvestment in existing IS.ARTICLE IN PRESSC.-S. Yu, Y.-H. Tao / Technovation 29 (2009) 92–109 99
SN was assessed by eight items adapted from the worksof 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 andDaniel (2004), and Driedonks et al. (2005). Respondentswere asked to express their degree of agreement with thefollowing statements using a seven-point Likert scaleranging from 1 (strongly disagree) to 7 (strongly agree):
13.
a majority of leading enterprises within the supplychain use e-marketplaces;14.
a majority of trading parties within the supply chainuse e-marketplaces;15.
a majority of peer competitors use e-marketplaces; 16. leading firms in the industry recognize that ane-marketplace can enhance firm competitiveness;
17. trading counterparts recognize that e-marketplaces canenhance firm competitiveness;
18. peer competitors recognize that e-marketplaces canenhance firm competitiveness;
19. the government actively promotes e-marketplaces; and 20. e-marketplace adoption is supported by governmentgrants.
Adapted from the works of Grover and Goslar (1993),Iacovou et al. (1995), Thong and Yap (1995), Thong(1999), Grewal et al. (2001), White and Daniel (2004),and Yu (2006), FC was assessed by 12 items, as listedbelow, on a seven-point Likert scale:
21.
a majority of data communication tasks are processedvia 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 interconnectedwith IS;
25. all trade processes are clear and distinct; 26. all trade processes are documented; 27. all questions regarding trade processes can be answeredusing the documentation;
28. all trade processes are easy to computerize; 29. the timing for locating/attracting prospective traders iscrucially important;
30. the timing for exchanging offerings with traders iscrucially important;
31. the timing for instant communication with traders iscrucially important; and
32. the timing for completing a transaction is cruciallyimportant.
Based on the studies of O’Callaghan et al. (1992), Groverand Goslar (1993), White and Daniel (2004), Ganeshet al. (2004), and Yu (2006), ICEC was operationalizedby asking the respondents the following five questions:
33.
your firm is strongly influenced by the industryenvironment in terms of high-velocity competitiverequirement;34.
your firm is strongly influenced by the industryenvironment in terms of supplier enablement;35.
your firm is strongly influenced by the industryenvironment in terms of buyer enablement;36.
your firm is strongly influenced by the industryenvironment in terms of channel power;37.
your firm is strongly influenced by the industryenvironment in terms of product power.In contrast to using a single question to assess attitudeas in most of the individual-level TAM literature, thiswork uses multiple items to operationalize business-level attitude due to business-level perception being acollective perception of the whole business’s decision-making members (Nelson and Quick, 2006). Moonet al. (2003) performed two group decision studies andempirically demonstrated that the decisions madeby groups whose members with and without priorindividual consideration of the problem exerted adifferent impact of the group decisions. The study ofMoon et al. implies that technology adoption con-siderations of key stakeholders may markedly influencethe 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 e-market-place adoption was operationalized by asking respon-dents the following four questions:
38.
the CEO strongly recognizes that e-marketplaces canenhance firm competitiveness;39.
the CEO has high awareness of e-marketplaces; 40. senior management strongly recognizes that e-market-places can enhance firm competitiveness; and
41. senior management has good awareness of e-market-places.
4.2. Sampling
Based on the above construct of operationalizationdrawn from the pertinent literature, the questionnairecomprises two sections. The first section contains 41questions assessed using a Likert-type scale ranging from1 (strongly disagree) to 7 (strongly agree), and collects theassessment of six constructs. The seven-point scale waschosen owing to its being more suitable for multi-variantanalysis than smaller ranges, such as a five-point scale. Thesecond section, containing nine questions listed in Appen-dix 1, gathers basic data on each respondent company andaims to determine whether or not the responding firm hasjoined an e-marketplace, whether or not the adopted firmsplan to continue using, switch, or stop using e-market-places, and whether non-adopting firms plan to adopte-marketplaces or not.Before officially issuing questionnaires, a pretest was
performed on scholars and participants of e-marketplacesto reword and refine the survey questions. Instead of
ARTICLE IN PRESS
Table 2
Respondent profile
Category Item Mean or frequency Std. dev. or %
All firms Number of e-marketplace adopters 94 46.53%
Number of e-marketplace non-adopters 108 53.47%
All firms Number of employees (person) 1069 2494
Capital (millions of NT$a) 2067 2902
Revenue (millions of NT$a) 7690 37,705
Industry type for all firms Chemistry, cement, petrochemistry 22 10.9%
Semiconductor 12 5.9%
Textile 14 6.9%
Optics, machinery, and metal 28 13.9%
Electronics and information 63 31.2%
Automobile 15 7.4%
Steel 12 5.9%
Medicine 5 2.5%
Food 7 3.5%
Others 24 11.9%
Firms that adopted e-marketplaces Number of employee (person) 1351 3131
Capital (millions of NT$a) 2462 3502
Revenue (millions of NT$a) 10,442 50,420
Annual membership fee (NT$a) 42,194 60,817
Firms that not adopted e-marketplaces Number of employee (person) 824 1743
Capital (millions of NT$a) 1577 1821
Revenue (millions of NT$a) 4334 7004
a1 US$A32.88 NT$ over the recent years 2004–2006.
C.-S. Yu, Y.-H. Tao / Technovation 29 (2009) 92–109100
mailing out the questionnaires, the pretest was conductedvia face-to-face interviews to ensure that all questions andterms used could be clearly understood by respondents.Like past business-level TAM studies (Amoako-Gyampahand Salam, 2004; Zain et al., 2005) and the prevailingbusiness-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 studyused the key informant method. To ensure the representa-tiveness of the respondents, clear and concise statementsdescribing the purpose of the research were provided at thebeginning of the questionnaire, and executives or managerswho are responsible for firm e-marketplace adoption wereinvited to complete the questionnaire. The survey wasissued to 1500 large-size firms randomly selected from theTop 5000 Company List published by China CreditInformation Service LTD (http://www.credit.com.tw/new-web/DB/index.htm).
After issuing two follow-up reminder mails, this studyreceived 295 questionnaires. Among 295 responses, 202were considered valid, representing a valid response rate of13.5%. Compared to survey return rates ranging from11.5% to 16.5% for similar empirical studies on Taiwanindustry during the recent years (Yu, 2006), a 13.5% validresponse rate generated from a 19.7% total response ratewas reasonable and compatible with recent surveys onTaiwanese firms. Table 2 briefly profiles the 202 firms asfollows: 94 respondents had used at least one e-market-place; these firms are highly profitable based on the ratio of
revenue over capital exceeding 300%; the percentages ofdifferent industry type also suitably reflect the currentindustry distribution of the electronic, information, optics,machinery, metal, chemistry, and semiconductor industriesin Taiwan. Moreover, the average annual membership feeof NT$ 42,194 (roughly US$1310) is not high for largefirms.
4.3. Validity and reliability
Based on the research structure in Fig. 3, it appearsreasonable to apply the structural equation model using asoftware such as LISREL for the hypothesis testing.However, there is a lack of sufficient theory-basedliterature investigating the antecedents of organizationalparticipation in B2B e-marketplaces as listed in Table 1.Particularly, this investigation presents a multi-modelresearch structure that is first drawn from TAM and IDTto investigate firm-level innovation technology adoption,namely e-marketplace. Therefore, this study approachesthe data analysis through factor analysis to determine theconstruct validity and regression analysis for hypothesistesting.The use of factor analysis is motivated by Yang and Yoo
(2004), who claimed that attitude is important in TAM buthas been ignored owing to the cognitive aspect that mattersin TAM not being distinguished from the affection aspectin previous studies. The regression method is used becauseit can not only use a limited number of predictor variables
ARTICLE IN PRESSC.-S. Yu, Y.-H. Tao / Technovation 29 (2009) 92–109 101
to systematically clarify the tendency of the responsevariable, but can also quantify the relationship betweendependent and independent variables and the explanatorypower of the entire model (Neter et al., 1999). This mayexplain why Davis, who invented TAM in 1986, has alwaysused the regression method to examine the hypothesesbased on TAM, extended TAM, TAM 2, and unified TAM(Davis, 1989, 1993; Davis et al., 1989; Venkatesh andDavis, 1996, 2000; Venkatesh et al., 2003).
Consequently, factor analysis using the principal com-ponent method with Varimax rotation was performed toverify whether or not the questionnaire items properly mapthe corresponding constructs. The criterion for each sortedquestion pertaining to each factor is that the eigenvaluemust exceed 1.0, and the corresponding intra-factorloading must exceed 0.6. After conducting the factoranalysis using the SPSS software, questions 4 and 12 wereremoved because they failed to meet the above-mentionedcriterion. Based on the results of factor analysis, there aretwo sub-constructs under PU and three sub-constructsunder FC. Each sub-construct name is given by reflectingthe context of the corresponding items as displayed inTable 3.
As listed in the last column of Table 3, the computedCronbach’s alpha coefficients for all dimensions exceed0.867, indicating high content consistency between thequestions relating to each of the constructs. Additionally,as displayed in Table 4, all of the inter-item correlationcoefficients under each construct are significant (po0.001),revealing that the content has reliable, convergent, anddiscriminating properties (Davis, 1989; Adams et al., 1992).This study also examined the correlation coefficientsamong constructs and found positive correlations amongthe constructs. This finding demonstrated that firm attitudeforms holistically rather than piecemeal, and indicatespotential overlap among some of the constructs.
5. Hypotheses test and data analysis
5.1. Hypothesis testing
Since regression is useful for forecasting the tendency ofresponse variables in a systematic fashion with a limitednumber of explanatory variables (Neter et al., 1999), thelinear regression method was used to verify hypotheses 1and 2, while the logistic regression method was applied toexamine hypotheses 3–7 due to their binary dependentvariables with only two possible outcomes, namely yes andno. The computed results are summarized in Table 5.
In model 1, the adjusted R2 demonstrates that 83.2%of firm attitude regarding e-marketplace adoption canbe explained, while PEOU strongly influences PU inrelation to business-level new technology adoption(p-valueo0.001). Additionally, the constructs of PU andSN considerably influence business attitude on e-market-place adoption (p-valueo0.001), while PEOU and FCstrongly influence business attitude toward e-marketplace
adoption (p-valueo0.01), and ICEC does not statisticallysignificantly affect business attitude toward e-marketplaceadoption. Thus, hypotheses 1a–1d and 2 are accepted, andonly hypothesis 1e is rejected.One worthy-mentioned issue is the multicollinearity
problem, which was not discussed in previous TAM studiesusing multiple regression method. Since PEOU signifi-cantly influences PU, when both PEOU and PU are used tomeasure the relationship with attitude in model 1, theproblem of multicollineartiy indeed needs to be checked upvia the tolerance and the VIF values. In fact, multi-collinearity is the property of every multiple regressionmodel, but the multicollinearity problem occurs only whenit significantly exceeds a certain level. After resetting theSPSS software with the statistics option of multicollinearitydiagnosis, the tolerance values are all greater than 0.2(0.321, 0.732, 0.338, 0.705, 0.624) and the VIF values are allsmaller than 5 (3.114, 1.366, 2.961, 1.418, 1.603) for all thefive independent variables, which, according to O’Brian(2007), are acceptable in practice.Model 2 examines the relationship between the business
decision regarding e-marketplace adoption and five con-structs of PU, PEOU, SN, FC, and ICEC, as well as therelationship between the business decision and attituderegarding e-marketplace adoption. The empirical resultsindicate that the hypotheses 3a, 3c, and 4 are accepted,whereas hypotheses 3b, 3d, and 3e are rejected. That is, thepositive relationship between the business attitude anddecision is very significant, and among the five constructsonly PU and SN significantly impact business decision one-marketplace adoption.Regarding model 3, the positive relationship between
business decision continuity and attitude is verified.That is, business adoption/non-adoption continuity isextremely significantly impacted by business attitude(p-valueo0.001). The decision continuity of the e-market-place adopting businesses is very significantly influenced byPU (p-valueo0.01), and is also strongly influenced by FCand ICEC (p-valueo0.05). Meanwhile, for non-adoptingbusinesses, the logistical regression results display thatnone of the five constructs considerably affect the decisioncontinuity of the non-adopting businesses. In this case,hypotheses 5a, 5d, 5e, and 7 are accepted, while hypotheses5b, 5c, and 6a–6e are rejected.From the above analysis, at the pre-decision stage this
study concluded that PU and SN are extremely importantfactors (p-valueso0.001), while PEOU and FC are veryimportant factors (p-valueso0.01) influencing businessattitudes. At the in-decision stage, this study verifies thatonly PU (p-valueso0.05) and SN (p-valueso0.01) sig-nificantly influence business attitudes, while businessattitude is an extremely significant factor (p-valueso0.001)influencing business decision. At the post-decision stage,none among PU, PEOU, SN, FC, and ICEC significantlyinfluence non-adopted business decision continuity aboutcontinuously not using e-marketplaces or planning to usee-marketplaces. However, PU very significantly influences
ARTICLE IN PRESS
Table 3
Factor analysis summary
Construct Named sub-construct Q# Factor loading Eigenvalue Cumulated variance (%) Cronbach a
PU Transaction functionality Q5 0.882 3.282 46.89 0.867
Q6 0.930
Q7 0.903
Q8 0.842
Competitive advantages Q1 0.936 2.585 83.82
Q2 0.958
Q3 0.830
PEOU Q10 0.936 2.557 85.23 0.913
Q11 0.926
Q9 0.907
SN Q13 0.905 3.394 79.79 0.872
Q14 0.862
Q15 0.830
Q16 0.888
Q17 0.926
Q18 0.864
Q19 0.832
Q20 0.749
FC Speed Q29 0.812 3.085 25.71 0.888
Q30 0.840
Q31 0.839
Q32 0.796
Standardization Q25 0.806 2.837 49.35
Q26 0.738
Q27 0.732
Q28 0.772
Computerization Q21 0.712 2.552 70.62
Q22 0.641
Q23 0.847
Q24 0.783
ICEC Q33 0.907 3.977 79.53 0.935
Q34 0.920
Q35 0.905
Q36 0.840
Q37 0.885
Attitude Q38 0.904 3.331 83.27 0.933
Q39 0.917
Q40 0.910
Q41 0.918
C.-S. Yu, Y.-H. Tao / Technovation 29 (2009) 92–109102
(p-valueso0.01) and FC and ICEC significantly influence(p-valueso0.05) adopted business decision continuityregarding whether to continue using current e-market-places, based on which the following three conclusionscan be reached. First, adopted businesses can maintaintheir original adoption decisions as long as they benefitfrom e-marketplaces adoption (i.e., the trading volume isgrowing due to using e-marketplaces, number of customersconducting transactions via e-marketplaces is growing,etc.). Second, the effect of PEOU is decreasing overtime since businesses are used to using e-marketplaces.Third, the effect of SN is weakening over time owing tothe increased first-hand experience of businesses withe-marketplaces.
Moreover, compared to the previous literature on B2Be-marketplace adoption, such as Yu (2006), which ob-served that external environmental pressures significantlyinfluence e-marketplace adoption, this study discoveredthat ICEC did not significantly influence business attitudetoward e-marketplace adoption during the pre-decisionstage, and also did not influence business decisionsregarding e-marketplace adoption during the in-decisionstage. During the post-decision stage, ICEC did notsignificantly influence non-adopting business decisioncontinuity, but ICEC significantly affected adopted busi-ness decision continuity (p-valueo0.05). This phenomenoncan be attributed to ICEC representing only industrycompetitive environment in this study, because other
ARTICLE IN PRESS
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*** 0.829*** 1 Q8
0.247** 0.260** 0.430** 0.702*** 0.754*** 0.680*** 1
Construct: PEOU
Q9
1 Q10
0.771*** 1 Q11
0.746*** 0.818*** 1
Construct: SN
Q13
1 Q14
0.818*** 1 Q15
0.768*** 0.725*** 1 Q16
0.715*** 0.806*** 0.831*** 1 Q17
0.507*** 0.356** 0.497*** 0.425** 1 Q18
0.489** 0.310** 0.407** 0.309** 0.763*** 1 Q19
0.464** 0.437** 0.440** 0.440** 0.615*** 0.710*** 1 Q20
0.422** 0.341** 0.258** 0.310** 0.288** 0.247** 0.247** 1
Construct: FC
Q21
1 Q22 Computerization
0.527*** 1 Q23
0.650*** 0.558*** 1 Q24
0.573*** 0.529*** 0.585*** 1
Q25
0.358** 0.201** 0.260** 0.229** 1 Q26 Standardization
0.397** 0.373** 0.416** 0.279** 0.679*** 1 Q27
0.387** 0.236** 0.326** 0.252** 0.522*** 0.546*** 1 Q28
0.297** 0.278** 0.314** 0.326** 0.661*** 0.649*** 0.586*** 1
Q29
0.324** 0.167* 0.348** 0.326** 0.215** 0.274** 0.329** 0.305** 1 Q30 Speed
0.394** 0.234** 0.293** 0.381** 0.175* 0.266** 0.259** 0.265** 0.668*** 1 Q31
0.352** 0.280** 0.302** 0.390** 0.213** 0.322** 0.302** 0.340** 0.560*** 0.600*** 1 Q32
0.215** 0.152* 0.223** 0.292** 0.161* 0.273** 0.330** 0.214** 0.685*** 0.595*** 0.547*** 1
Construct: ICEC
Q33
1 Q34
0.833*** 1 Q35
0.685*** 0.669*** 1 Q36
0.738*** 0.725*** 0.744*** 1 Q37
0.709*** 0.684*** 0.625*** 0.622*** 1
Construct: Attitude
Q38
1 Q39
0.771*** 1 Q40
0.746*** 0.818*** 1 Q41
0.614*** 0.644*** 0.699*** 1
*po0.05, **po0.01, ***po0.001.
C.-S. Yu, Y.-H. Tao / Technovation 29 (2009) 92–109 103
external factors such as the influence of government,the influence of leading or powerful stakeholders in theindustry, and the like are covered by SN in this study.
Besides, the survey units were small and medium compa-nies in Yu (2006) while this work sampled large companies.Therefore, instead of saying that large firms are less
ARTIC
LEIN
PRES
STable 5
Summary of regression test
Model Dependent variables Independent variables Standardized beta
value
t-Value Hypothesis Adjusted R2 (F-value)
Results of hypotheses 1–2
1 Attitude PU 0.165 4.588*** H1a accepted 0.832 (191***)
PEOU 0.110 3.142** H1b accepted
SN 0.713 14.106*** H1c accepted
FC 0.132 3.164** H1d accepted
ICEC 0.022 0.657 H1e rejected
PU PEOU 0.486 7.713*** H2 accepted 0.233 (59***)
Beta value Wald chi-square Hypothesis Model summary
Results of hypotheses 3–7
2 Adoption decision PU 0.480 3.945* H3a accepted X2 (df ¼ 5) ¼ 30.844, p-value ¼ 0.008,
�2log likelihood ¼ 237.767, overall correct
classification rate ¼ 67.5% (66.7% for adopters
and 68.3% for non-adopters)
PEOU 0.344 2.871 H3b rejected
SN 1.040 10.389** H3c accepted
FC 0.302 2.676 H3d rejected
ICEC 0.097 0.127 H3e rejected
Adoption decision Attitude 0.447 9.658*** H4 accepted X2 (df ¼ 1) ¼ 15.188, p-value ¼ 0.000,
�2log likelihood ¼ 226.855, overall correct
classification rate ¼ 83.1% (84.2% for adopters
and 82.1% for non-adopters)
3 Adopted firms’
decision continuity
PU 3.296 7.244** H5a accepted X2 (df ¼ 5) ¼ 45.650, p-value ¼ 0.002,
�2log likelihood ¼ 274.834, overall correct
classification rate ¼ 71.65% (70.1% for staying
in and 84.5% for exiting)
PEOU 1.407 3.228 H5b rejected
SN 1.185 1.769 H5c rejected
FC 1.812 4.668* H5d accepted
ICEC 1.220 4.023* H5e accepted
Non-adopted firms’
decision continuity
PU 0.299 0.379 H6a rejected X2 (df ¼ 5) ¼ 1.154, p-value ¼ 0.949, �2log
likelihood ¼ 89.612, overall correct classification
rate ¼ 55.5% (62% for planning to continuously
not adopt e-marketplaces and 40% for planning
to adopt)
PEOU 0.101 0.101 H6b rejected
SN 0.093 0.060 H6c rejected
FC 0.243 0.283 H6d rejected
ICEC 0.146 0.178 H6e rejected
Decision continuity Attitude 0.631 9.961*** H7 accepted X2 (df ¼ 1) ¼ 12.261, p-value ¼ 0.000, �2log
likelihood ¼ 163.422, overall correct
classification rate ¼ 76.53% (71.2% for adopted
firms and 81.3% for non-adopted firms)
*Significant at 0.05 level, **significant at 0.01 level, ***significant at 0.001 level.
C.-S
.Y
u,
Y.-H
.T
ao
/T
echn
ova
tion
29
(2
00
9)
92
–1
09
104
ARTICLE IN PRESS
Table 6
t-Test summary
Adopted firm Non-adopted firm t-Value
Meana Std. dev. Meana Std. dev.
PU 5.22 0.73 4.86 0.87 3.190**
Transaction functionality 5.34 0.90 5.15 0.93 1.480
Competitive advantages 5.07 0.97 4.47 1.17 3.938***
PEOU 3.25 0.93 2.91 1.00 2.971**
SN 5.23 0.77 4.64 1.03 4.567***
FC 5.58 0.73 5.34 0.74 2.246*
Speed 5.85 0.72 5.79 0.87 0.536
Standardization 5.54 0.84 5.29 0.97 1.998*
Computerization 5.28 1.00 4.87 1.04 2.854**
ICEC 4.981 1.00 4.71 0.99 1.918
*Significant at 0.05 level, **very significant at 0.01 level, ***extremely
significant at 0.001 level.aAverage scores based on the seven-point Likert scale with 1 for
strongly disagree and 7 for strongly agree.
C.-S. Yu, Y.-H. Tao / Technovation 29 (2009) 92–109 105
impacted by the industry than small and medium firms, itmay be more reasonable to say that small and mediumbusinesses are more easily influenced by industry.
5.2. Adopted vs. non-adopted firm analysis
As Karahanna et al. (1999) noted, TAM studies typicallyexamine the attitudes of all respondents regarding the ITadoption, but few studies have examined adopters (currentcustomers) and non-adopters (called potential customers),respectively. Ignoring the examination of the difference inview of both adopters and non-adopters toward innovationtechnology adoption may lead to certain valuable cluesrelated to new technology adoption being lost. By using thet-test to examine the mean difference between adopters andnon-adopters in terms of the constructs and sub-constructs,Table 6 shows the mean and standard deviation of theseconstructs for both adopters and non-adopters.
For the macro-level of the five constructs, Table 6 revealsthat the scores on PEOU are especially low (average scoreless 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 thefact that although e-marketplaces are Internet-based opensystems, their system interfaces and functionalities remaininsufficiently friendly, which may partly explain why theadoption rate of e-marketplaces in Taiwan remains belowexpectations (Yu, 2006). From the micro-level perspective,the figures regarding two sub-constructs of transactionfunctionality and competitive advantages under the PU inTable 6 demonstrate that transaction functionality doesnot impact business decisions on e-marketplace adoption,but competitive advantages are a crucial factor and lead toPU significantly influencing e-marketplace adoption.
6. Implications for theory and business
This study surveyed 202 large Taiwanese firms and thusderived four theoretical implications which paves the wayfor advancing current knowledge of the business-level newtechnology adoption. Meanwhile, some business implica-tions are briefly described, which may be of interest tothose who are interested in effectively selling innovationIT/IS to enterprises.
6.1. Theoretical implications
Compared to individual-level technology adoptionstudies in which the variance explained by TAM or itsvariations is generally less than 40% (Hung et al., 2005),this empirical study has demonstrated that TAM possessesmore powerful explanatory ability in B2B e-marketplaceadoption (in which adjusted R2 is around 83%, as shown inTable 5). Accordingly, the first theoretical implicationdrawn from this empirical study is that TAM is not onlyuseful in foreseeing individual-level new technology adop-tion but also works well in predicting business-level
technology adoption (i.e., e-marketplace). This result maybe attributed to collective business decisions involvingmore rationality and a longer decision time than single-person decisions. However, at the post-decision stage, theresults of this study indicate that TAM cannot effectivelyexplain business decision continuity following originaladoption/non-adoption. Accordingly, to include othertheories (i.e., expected-performance theory), research intothe structure is required to understand what influencesbusiness-level decision continuity. Accordingly, the secondtheoretical implication is that TAM is useful only inexplaining business attitude and decision regarding newtechnology adoption, but cannot predict business decisioncontinuity after adoption/non-adoption.The study of Moon et al. (2003) argued that the decisions
made by groups whose members have and have not givenprior individual consideration to the problem exert adifferent impact on group decisions. Similarly, manyorganizational behavior studies contend that organiza-tional decision behavior has not only inherited the rationaland irrational components of individual decisions butmust also satisfy the concerns of multi-dimensionalstakeholders (Nelson and Quick, 2006). Following theconcept behind these studies, this research surveys 202large Taiwanese firms regarding e-marketplace adoption.The empirical results demonstrate that enterprise beliefssignificantly impact enterprise attitudes during the pre-adoption, and enterprise attitudes significantly impactenterprise decision and decision continuity during the in-decision and post-decision stages. Consequently, the thirdtheoretical implication is that business-level attituderegarding innovation technology adoption during thepre-adoption can be effectively explained by businessbeliefs, and organization-level decision and decision con-tinuity during the in-decision and post-decision arestrongly influenced by business-level attitudes. Certainly,
ARTICLE IN PRESSC.-S. Yu, Y.-H. Tao / Technovation 29 (2009) 92–109106
more empirical studies to examine the presented researchmodel are necessary.
Looking at Tables 5 and 6, the figures have shown thatSN significantly shapes business attitude before thedecision and very strongly influences business decisionduring the decision itself. That is, potential adopters mayacquire information from early adopters in addition tobeing influenced by adopters with positive experience(Rogers, 2003) or being forced by other importantstakeholders in the industry (Driedonks et al., 2005). Asargued by Frambach and Schillewaert (2002) and Moonet al. (2003), organizational-level attitude is the collectiveattitude of organization decision makers, and organiza-tional decision is made by a group of members who areinfluenced by their respective SN considerations of theproblem just as for individual-level new technologyadoption. Accordingly, this study confirmed that SN isnot only a crucial factor influencing individual-leveltechnology adoption, but also influences business-leveltechnology adoption, which leads to the fourth theoreticalimplication.
6.2. Business implications
Recently, customer relationship management has re-ceived increasing attention, with a particular emphasis onthe importance of understanding old customers. Althoughthis work did not precisely identify the influences on thedecision continuity of non-adopting firms at the post-decision stages, it identified the influences on the continueduse of e-marketplaces by adopting firms. As shown inTable 5, three constructs of PU, FC, and ICEC signifi-cantly 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 highusefulness expectations and low e-readiness and industrycompetitive environment. Meanwhile, some adopting firmswith lower royalty and satisfaction can be identified, andthen marketers should devise suitable service strategies toupgrade their satisfaction and royalty.
Furthermore, figures listed in Table 5 showed thatattitude still significantly influences decisions of adoptingfirms to continue using current e-marketplaces or thoseof non-adopting firms to continue rejecting the use ofe-marketplaces during the post-decision stage. Therefore, ifan e-marketplace wishes to alter the original decisions ofnon-adopting firm, the marketers should first change firmattitudes. Similarly, if an e-marketplace wishes to retaincurrent adopting firms, the original attitudes of thosefirms must be maintained and better enhanced to raisetheir royalty and usage. Firms will never change theiroriginal attitudes without some relevant motivation. Sincedeterminants influencing business attitudes regardinge-marketplace adoption have been identified in this work,e-marketplace managers may feature their research anddevelopment, marketing, and service strategies based onthese findings, and thus reverse the attitudes of non-
adopting firms as well as strengthen the beliefs of adoptingfirms.Table 5 demonstrates that PU, PEOU, SN, and FC
significantly influence firm attitudes, while only PU and SNstrongly influence business decisions. Hence, new technol-ogy marketers can prioritize their strategic focus subject todifferent stages such as selling innovation IT/IS topotential business-level customers at the pre-adoptionstage, enhancing the willingness of existing users tocontinuously use their IT/IS at the post-adoption stage.This implies that the theme of marketing programs shouldbe adjusted as a subject of different stages to the users orprospective users, and the service programs must beeffectively differentiated subject to different stages.Moreover, combining Tables 5 and 6 together, this study
found that SN exerts a much more powerful influence thanthe others. That is, the greater the effect of SN, the higherthe likelihood of the firms adopting e-marketplaces. As aresult, e-marketplace marketers may invite importantstakeholders in the industry or leading/famous businessesto execute the testimonial marketing events to attract non-adopting firms to become adopters. Based on Table 6, thestandardization and computerization are enablers throughwhich FC significantly impacts business adoption ofe-marketplaces. Consequently, marketers may select busi-nesses with higher levels of workflow computerization and/or standardization as priority customers (the most pro-spective users).Referring to previous TAM-based business-level studies
(Zain et al., 2005; Amoako-Gyampah and Salam, 2004),and Zain et al. (2005) concluded that PEOU affects onlyfirm attitude and not PU, while Amoako-Gyampah andSalam (2004) discovered that PEOU significantly influencesPU while only PU influences firm attitude. Compared toZain et al. (2005) and Amoako-Gyampah and Salam(2004), this study found that PEOU significantly impactsPU, while both PU and PEOU significantly impactcompany attitude. Regarding the relationship betweenPEOU and PU, this work supported Amoako-Gyampahand Salam (2004) instead of Zain et al. (2005), which maybe attributed to the research object in Amoako-Gyampahand Salam (2004) of ERP system or inter-organizationalIS, which is more complex than the desktop computer usedas the research object in Zain et al. (2005). Accordingly,another business implication may result as follows: theinfluence of PEOU on PU decreases with decreasingcomplexity of the new IT/IS-based products, the influenceof PU increases with increasing complexity of the new IT/IS-based products. That is, the influence of PEOU on PUand business attitude and the influence of PU on businessattitude are changeable and rely on the complexity of IT/ISitself.Building on the above discussion, when developing/
launching an innovative IT/IS, the businesses shouldprioritize their marketing efforts on SN, focus theirresearch and development efforts on PU and PEOU, andtarget their sales efforts on enterprises with higher levels of
ARTICLE IN PRESSC.-S. Yu, Y.-H. Tao / Technovation 29 (2009) 92–109 107
workflow computerization and standardization. Anyhow,the above discussion simply demonstrates some businessimplication drawn from this empirical study. To providebusiness with more useful clues, further elaborate researchis needed.
7. Concluding remarks
Like any study, this work naturally suffers from certainlimitations. First, due to the increasing number of highereducation institutes over the last 15 years, from less than 30initially to over 160 now, as well as to business andindustry gradually moving to China (Tao et al., 2007), thisstudy had practical difficulties in achieving a high responserate in the survey of Taiwanese firms. Future studies couldovercome this problem by conducting qualitative casestudies via in-depth face-to-face interviews to collect data,which could reexamine the business-level new IT/ISadoption model presented in this study. Second, this initialfirm-level TAM study used regression method to analyzethe collected data as Davis (1989) did, which is adequatefor simple path models like the ones proposed in this study.However, for future insightful firm-level TAM studies, theStructured Equation Modelling (SEM)-base method isrecommended for more sophisticated path models. Third,this was not a longitudinal study on examining the samesample from pre-adoption to the post-adoption stages. Toverify the suitability of IDT for explaining business-levelinnovation diffusion, studies on a set of same samples fromthe pre-adoption, through the in-adoption, and finally tothe post-adoption stages are necessary. Finally, since thesample was obtained only from Taiwanese enterprises ande-marketplace is an innovation IT/IS highlighting on theInternet context, caution is necessary in generalizing themethodology and findings to other technologies or othercountries with different cultures or industrial structure.
However, by replacing the individual with the organiza-tion as an analysis unit in the measurement, this studyconfirms that TAM can effectively explain new technologyadoption by enterprises. Compared to the vast individual-level TAM literature, business-level TAM literature isrelatively scarce and has devoted insufficient attention toexploring the antecedents and consequences of business-level attitude and decision on new IT/IS adoption. There-fore, an important contribution of this study is to fill thisgap, build links between TAM and business-level technol-ogy adoption, and pave the theoretical ground foradvancing current understanding of business-level innova-tion technology adoption.
Besides, since e-marketplace is a web-based IT/IS,findings of this empirical study may provide interestedparties with some clues about how to promote and sellweb-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 theadoption of other business-level technology adoption.Finally, this investigation merely represents a preliminary
work aimed at improving the understanding of business-level innovation technology adoption from the perspectiveof behavioral theories. Further research is definitelyrequired to verify and enhance the validity and general-ization of the methodology used in this study.
Appendix 1. The second section of the questionnaire
The following data will be used confidently for academicresearch only.
Q42.
Has your company adopted any e-marketplaceyet? & Yes & NoQ43.
What is the employee size of your company? ______PersonsQ44.
What is the approximate capital of your company?US$ ______Q45.
What is the approximate annual revenue of yourcompany? US$ ______Q46.
What is the industry type of your company? ______ Q47. What is your position title in this company? ______ Q48. The department that you work is& Procurement Department & SalesDepartment & Others ______If your firm has adopted an e-marketplace,please answer Q49. Otherwise, please answerQ50.
Q49.
Will your company continue to use the existing e-marketplace?& Yes & No
Q50. Does your company plan to adopt an e-marketplacewithin a year?& Yes & No
References
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.
Ajzen, I., Fishbein, M., 1980. Understanding Attitudes and Predicting
Social Behavior. Prentice-Hall, Englewood Cliffs, NJ.
Amoako-Gyampah, K., Salam, A.F., 2004. An extension of the
technology acceptance model in an ERP implementation environment.
Information and Management 41 (2), 731–745.
Angeles, R., 2000. Revisiting the role of Internet-EDI in the current
electronic commerce scene. Logistics Information Management 13 (1),
45–50.
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.
Bakos, J.Y., 1991. A strategic analysis of electronic marketplaces. MIS
Quarterly 15 (3), 295–310.
Bakos, J.Y., 1997. Reducing buyer search costs: implications for electronic
marketplaces. Management Science 43 (12), 1676–1692.
Benbasat, I., Barki, H., 2007. Quo vadis, TAM? Journal of the
Association for Information Systems 8 (4), 211–218.
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.
ARTICLE IN PRESSC.-S. Yu, Y.-H. Tao / Technovation 29 (2009) 92–109108
Bhatacherjee, A., 2001. Understanding information system continuance:
an expectation-confirmation model. MIS Quarterly 25 (3), 351–370.
Chau, P.Y.K., 2001. Inhibitors to EDI adoption in small businesses: an
empirical investigation. Journal of Electronic Commerce Research 2
(2), 78–88.
Chau, P.Y.K., Tam, K.Y., 2000. Organizational adoption of open
systems: a technology-push, technology-pull perspective. Information
and Management 37 (2), 229–239.
Cooper, R.B., Zmud, R.W., 1990. Information technology implementa-
tion research: a technology diffusion approach. Management Science
36 (2), 123–139.
Davis, F.D., 1989. Perceived usefulness, perceived ease of use, and user of
acceptance of information technology. MIS Quarterly 13 (3), 319–340.
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.
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.
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.
Fabiani, S., Fabiano, S., Trento, S., 2005. ICT adoption in Italian
manufacturing: firm-level evidence. Industrial and Corporate Change
14 (2), 225–249.
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.
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.
Gatignon, H., Robertson, T.S., 1989. Technology diffusion: an empirical
test of competitive effects. Journal of Marketing 53 (1), 35–49.
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.
Gottschalk, P., Abrahamsen, A.F., 2002. Plans to utilize electronic
marketplaces: the case of B2B procurement markets in Norway.
Industrial Management and Data Systems 102 (5), 325–331.
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.
Grover, V., Goslar, M.D., 1993. The initiation, adoption, and implemen-
tation of telecommunications technologies in US organization. Journal
of Management Information Systems 10 (1), 141–163.
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.
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.
Hartwick, J., Barki, H., 1994. Explaining the role of user participation in
information system use. Management Science 40, 440–465.
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.
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.
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.
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.
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.
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.
King, W.R., He, J., 2006. A meta-analysis of the technology acceptance
model. Information and Management 43 (2), 740–755.
Lai, W.S., 1998. Studying the decision factors and execution status of
enterprises in adopting e-store strategies. Master Thesis, Da-Ye
University, Taiwan.
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.
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.
Lin, H.F., Lin, S.M., 2008. Determinants of e-business diffusion: a test of
the technology diffusion perspective. Technovation 28 (1), 135–145.
Lucchetti, R., Sterlacchini, A., 2004. The adoption of ICT among SMEs:
evidence from an Italian Survey. Journal of Small Business 23 (2), 151–168.
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.
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.
McFarlan, F.W., 1984. Information technology changes the way you
compete. Harvard Business Review 62 (3), 98–103.
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.
Nelson, D.L., Quick, J.C., 2006. Organizational Behavior: Foundations,
Reality and Challenges. South-Western, Nashville, Tennessee.
Neter, J., Kutner, M.H., Nachtsheim, C.J., Wasserman, W., 1999. Applied
Linear Regression Models, third ed. McGraw-Hill, Taipei, Taiwan.
O’Brian, R.M., 2007. A caution regarding rules of thumb for variance
inflation factors. Quality and Quantity 41 (5), 673–690.
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.
Premkumar, G., Roberts, M., 1995. Adoption of computer aided software
engineering (CASE) technology: an innovation adoption perspective.
Data Base Advances 26 (2), 105–124.
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.
Rogers, E.M., 2003. Diffusion of Innovations, fifth ed. Free Press,
New York.
Schepers, J., Wetzels, M., 2007. A meta-analysis of the technology
acceptance model: investigating subjective norm and moderation
effects. Information Management 44, 90–103.
Stockdale, R., Standing, C., 2004. Benefits and barriers of electronic
marketplace participation: an SME perspective. Journal of Enterprises
Information Management 17 (4), 301–311.
Strader, T.J., Shaw, M.J., 1999. Consumer cost differences for traditional
and Internet markets. Internet Research 9 (2), 82–92.
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.
Thong, J.Y.L., Yap, C.S., 1995. CEO characteristics, organizational
characteristics and information technology adoption in small business.
Omega 23 (4), 429–442.
Thong, J.Y.L., 1999. An integrated model of information systems
adoption in small business. Journal of Management Information
Systems 15 (4), 187–214.
ARTICLE IN PRESSC.-S. Yu, Y.-H. Tao / Technovation 29 (2009) 92–109 109
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.
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.
Venkatesh, V., Davis, F.D., 2000. A theoretical extension of the
technology acceptance model: four longitudinal field studies. Manage-
ment Science 46 (2), 186–204.
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.
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.
White, A., Daniel, E., 2004. Electronic marketplaces: an empirical study in
the UK healthcare sector. International Journal of Electronic Business
2 (6), 603–623.
Yang, H.D., Yoo, Y., 2004. It’s all about attitude: revisiting the
technology acceptance model. Decision Support Systems 28 (1), 19–31.
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.
Yu, C.S., 2006. Exploring influences on Taiwanese SMEs e-marketplace
adoption decision. Journal of Global Information Technology
Management 9 (2), 5–21.
Zain, M., Rose, R.C., Abdullah, I., Masrom, M., 2005. The relationship
between information technology acceptance and organizational agility
in Malaysia. Information and Management 42, 829–839.
Zhu, K., 2004. The complementarily of information technology infra-
structure and e-commerce capability: a resource-based assessment of
their business value. Journal of Management Information Systems 21
(1), 167–202.
Zhu, K., Kraemer, K.L., Gurbaxani, V., Xu, S.X., 2006. Migration to
open-standard inter-organizational systems network effects, switching
costs, and path dependency. MIS Quarterly 30 (4), 515–538.
Chian-Son Yu is a professor in the Institute of Information
Management and Dean of School of Liberal Education at Shih
Chien University, Taipei, Taiwan. He received Ph.D. degree from
National Chiao Tung University, Hsinchu, Taiwan. His work
has been published in over 30 articles in scholarly journals,
including Journal of Electronic Commerce Research, Interna-
tional Journal of Global Information Technology, the Service
Industries Journal, Computers and Operations Research, Elec-
tronic Commerce Studies, European Journal of Operational
Research, Fuzzy Sets and Systems, Industrial Management and
Data Systems, International Journal of Production Research,
Management and System, Pan-Pacific Management Review, and
Journal of Information Management. He has also presented
papers at over 60 national and international conferences.
Yu-Hui Tao is a professor in the Department of Information
Management and Director of Extension Education Center at
National University of Kaohsiung, Kaohsiung, Taiwan. He
received Ph.D. degree from Ohio State University, Columbus,
Ohio, US. His work has been published in over 25 articles in
scholarly journals, including Computer and Human Behavior,
Computer and Education, Industrial Marketing Management,
Internet Research, Computers in Industry, International Journal
of Electronic Business Management, Journal of Information
Management, International Journal of Information Manage-
ment, Intelligent Data Analysis, and IIE Transactions. He has
also presented papers at over 45 national and international
conferences.