|
COPYRIGHT 2001 Institute for Operations Research and the Management Sciences
This paper is the first test of a parsimonious model that posits three factors as determinants of the adoption of electronic data interchange (EDI): readiness, perceived benefits, and external pressure. To construct the model, we identified and organized the factors that were found to be influential in prior EDI research. By testing all these factors together in one model, we are able to investigate their relative contributions to EDI adoption decisions. Senior purchasing managers, chosen for their experience with EDI and proximity to the EDI adoption decision, were surveyed and their responses analyzed using structural equation modeling. All three determinants were found to be significant predictors of intent to adopt EDI, with external pressure and readiness being considerably more important than perceived benefits. We show that the constructs in this model can be categorized into three levels: technological, organizational, and interorganizational. We hypothesize that these categories of influence will also be determinants of the adoption of other emerging forms of interorganizational systems (IOS). (1)
(Electronic Data Interchange; Electronic Commerce; Interorganizational Systems; Adoption of IT; Empirical Research; Partial Least Squares)
1. Introduction
According to the Economist (February 26, 2000), it is estimated that business-to-business (B2B) transactions will be more than 80% of the expected $3 trillion electronic commerce (EC) market by 2003. As B2B EC gains prominence, electronic data interchange (EDI) will remain an important enabling technology. To illustrate, the market for EDI software, products, and consulting services is predicted to grow from $800 million in 1997 to $2 billion annually in 2001 (Densmore 1998). Given the ongoing importance of EDI, the objective of this study is to test a parsimonious predictive model that posits three factors as determinants of the adoption of electronic data interchange (EDI): readiness, perceived benefits, and external pressure. To construct the model, we identified and organized the factors that were found to be influential in prior EDI research. By testing all these factors together in one model, we are able to investigate their relative contributions to EDI adoption decisions. We believe that this model can be generalized to other interorganizational information technology (IT) innovations.
Although 95% of Fortune 1000 firms have implemented EDI, only 2% of the remaining 6 million businesses in the U.S. have done so (Densmore 1998). Though the largest firms have aggressively encouraged EDI adoption, they have, on average, been able to motivate only 20% of their partners to adopt. The remainder, often comprised of small- and medium-sized enterprises (SMEs), have resisted adopting EDI for reasons that are not fully understood (Bouchard 1993, Hart and Saunders 1997). Given the ongoing importance of EDI, it is important to understand how to improve EDI adoption rates by developing and testing a parsimonious model that includes the key success factors identified in earlier work. Such work can also serve as the theoretical and empirical basis for research on other forms of interorganizational systems (IOS), such as business-to-business electronic commerce exchanges.
The rest of the paper is presented as follows. Section 1.1 reviews prior research on EDI. Section 2 describes the theoretical framework used in this paper. The research methodology is described in [section] 3, followed by the results in [section] 4. The paper concludes in [section] 5.
1.1. Prior Research on EDI
EDI promises many benefits, ranging from modest (reduced communication and administration costs and improved accuracy) to transformative (enabling business process reengineering or supporting industry value chain integration initiatives such as just-in-time inventory, continuous replenishment, and quick response retailing). Because of these potential benefits, EDI has been extensively studied using several theoretical perspectives.
A fundamental approach for the study of the adoption of new technologies is the diffusion of innovations (DOI) (Tornatzky and Klein 1982, Rogers 1995), which has been, either explicitly or implicitly, a foundation for much of EDI research (e.g., O'Callaghan et al. 1992, Premkumar et al. 1994, Teo et al. 1995). The focus of DOI research is on the "perceived characteristics of the innovation" that either encourage (e.g., relative advantage) or inhibit (e.g., complexity) adoption. For example, O'Callaghan et al. (1992) examined independent property and casualty insurance agents and found that relative advantage was a predictor of intent to adopt, as well as a differentiator between adopters and nonadopters. Likewise, in a survey of EDI adopters, Premkumar et al. (1994) found that relative advantage and compatibility are predictors of the extent of "adaptation"--the degree of EDI usage in its first application (operationalized as either purchase orders or invoices). Teo et al. (1995) used innovation diffusion theory to predict intent to adopt financial EDI in Singapore. Their findings show that complexity is a strong inhibitor of intent to adopt, as is their measure of the perceived risks of adopting.
Because the DOI-based research is focused on the perceived characteristics of the particular technology, we label this perspective "technological." While the technological perspective afforded by DOI undoubtedly explains a portion of the EDI adoption decision, it is primarily based on individual-level adoption decisions. However, EDI adoption is almost always an organizational-level decision executed in an interorganizational context; therefore, there are clearly aspects of the EDI adoption decision that are not captured by looking solely at (perceptions of) the technology of EDI. Thus, much of the research on EDI has taken an "organizational" approach, focusing on organizational characteristics as well as the inherent attributes of EDI technology. Although there is obvious overlap between the technological and the organizational perspectives, in light of the fact that perceived attributes of the technology are considered relative to the adopting organization, these two approaches are conceptually distinct in that they focus on different units of analysis: technologies versus organizations.
Organizational adoption of a technological innovation can be positioned within a much larger body of innovation research conducted by economists, technologists, and sociologists (see Gopalakrishnan and Damanpour 1997 for a comprehensive literature review). Within the sociologists group, the process view of innovation (or adoption of innovations) treats all innovations as equivalent units of analysis, and thus does not differentiate among different innovations with different attributes. Conversely, IS research can largely be classified into the variance sociologists group, and has focused on the innovation level of analysis and the development of "middle-range" theories of innovation (Gopalakrishnan and Damanpour 1997). Such theories focus on the attributes of the innovation and propose relationships between these attributes and the antecedents and consequences of adoption, acknowledging that some attributes of a particular technology will vary across organizations (such as compatibility).
Grover (1993), taking a comprehensive "bottom-up" approach, empirically identified five factors that statistically discriminated between firms that have and have not adopted EDI: (i) proactive technological organization, (ii) internal push, (iii) market assessment, (iv) competitive need, and (v) impediments. Reich and Benbasat (1990) examined the adoption of customer-oriented strategic systems, finding that adoption was related to customer awareness of need and support. Rogers (1995) examines the factors leading to organizational innovativeness, which include, among others, organizational slack and size. (Because this model focuses on the overall innovativeness of an organization--i.e., the process approach to innovation--rather than the adoption of a particular technology, it does not provide a testable model of EDI adoption.) The size and slack factors are one possible explanation for the greater rate of EDI adoption among very large (e.g., Fortune 1000) firms, as organization size has consistently been recognized as a driver of organizational innovation (see Damanpour 1992, for a meta-analysis).
Because adoption of EDI requires coordination between at least two organizations, the relationship between the organization and its prospective trading partner(s) becomes salient. In the best-case scenario, both firms agree that adoption is in their best interest. EDI is an example of a technology with positive externalities or network effects; thus, the actions of one firm will depend on (its perception of) the collective actions of other firms (i.e., are there enough firms adopting this technology to make our adoption worthwhile?). Collective actions and technology have been studied within a number of disciplines; Bouchard (1993) labels this collected work "critical mass theory." However, the positive benefits of having a critical mass of firms adopting the same technology is only one aspect of interorganizational relationships and EDI adoption. Another significant factor is enacted power, such as when one organization "encourages" or coerces its trading partners to adopt EDI. In the context of EDI adoption, we characterize factors relating to the actions of other organizations as belonging to the "interorganizational" level.
Recent EDI research has incorporated both interorganizational and organizational factors with somewhat mixed findings. Saunders and Clark (1992) examined the impact of perceived benefits and perceived costs (both technological factors), as well as dependency and trust (interorganizational factors) on intent to adopt EDI. They find that perceived costs reduce intent to adopt as does, somewhat surprisingly, trust. Bouchard (1993) found that DOI factors were insignificant in the EDI adoption decision, whereas the use or requirement of EDI by major business partners were the key drivers of the adoption decision. Premkumar and Ramamurthy (1995) found that the technological factor internal need (akin to perceived benefits) and the organizational factor top-management support, as well as the interorganizational factors competitive pressure and exercised power, influence whether a firm's EDI adoption decision is proactive or reactive. Iacovou et al. (1995) hypothesized a model that includes three factors as determinants of EDI adoption and impact in SMEs: perceived benefits (technological), organizational readiness (organizational), and external pressure (interorganizational). Proposing a high-low dichotomy for each of these factors led the authors to develop a 2 x 2 x 2 classification of firms; however, this model has not been empirically tested outside the small sample of firms that led to its genesis....
Read the full article for free courtesy of your local library.
|
Bookmark this article
Print this article
Link to this article
Email this article
Digg It!
Add to del.icio.us
RSS
