زبان مدل سازی یکپارچه (UML) پذیرش فن آوری اطلاعات و ارتباطات - مدل جامع نگر از دیدگاه قابلیت های سازمانی

This study develops an integrated research model to examine various factors affecting the IT adoption in the context of the Unified Modeling Language (UML). UML is one type of business process modeling techniques, which in turn is a key aspect of the business process reengineering. The proposed research model is based on IT adoption framework and organizational culture theory. The model identifies fourteen variables, covering seven broad categories (IT characteristics, organization technology, environment, organization structure, organization process, organization culture, and project culture) that could potentially impact UML adoption in organizations. This comprehensive conceptual model is further validated by survey data collected from 251 North American organizations across five different industries. Our results support the proposed conceptualization and shed new light on the key factors associated with firms' adoption of UML technologies. Theoretical and managerial implications of the findings are discussed.

The significance of understanding Information Technology (IT) adoption is well documented [1]. Considerable scholarly research has focused on investigating the impact of one or several of these factors (e.g., IT characteristics, organization technology, environment, organization structure, organization process, and organization culture) in different environmental settings [32], [57], [58] and [76]. Extant research has recognized the importance of technological, organizational, and environmental factors (TOE framework) in influencing IT adoption [8], [14], [80] and [83]. In spite of prior research that has found strong empirical support for the TOE framework, much fruitful theoretical work remains to be conducted. For instance, previous studies have proposed that IT adoption is affected by organization structure [14], [17] and [60], organization process [12], [14], [56] and [76] and organization culture as well as project culture [52], [60], [80] and [84]. In spite of extensive prior studies, there is a lack of comprehensive and integrative understanding, from the organizational culture perspective, on the IT adoption process, which is crucial for both practitioners and researchers in terms of generating deeper understanding of IT adoption. To fill this void, we extend the TOE framework in our research to incorporate organizational culture theory such as organization structure, organization process, organization culture, and project culture. Thus, practitioners may benefit from the holistic analysis of the determinants of IT adoption, and managers interested in introducing new technologies may be able to understand and act more effectively in terms of how to better facilitate IT adoption. The objective of this paper is therefore to identify a holistic IT adoption model to investigate three research questions on how various organizational factors will impact the IT adoption process. The purpose of this study is threefold. First, it seeks to investigate whether the technological, organizational, and environmental antecedents tailored for a specific context affect IT adoption. Second, it aims to explore whether various organizational idiosyncratic factors (organization structure, organization process and organization culture) determine their IT adoption. Third, it attempts to show the superiority of the holistic IT adoption model as compared to a traditional TOE framework. These research questions are examined in the context of Unified Modeling Language (UML) adoption using survey data collected in the United States across five industries. UML is a visual and graphical modeling language and has been increasingly used in the past decade in software engineering and e-commerce [1], enterprise modeling, business engineering, process analysis and system configuration [81]. The adoption of UML in organizational computing represents a major change in information systems development and implementation [30] and [78]. Despite the perceived benefits and its promotion by many industry leaders and the Object Management Group (OMG), the adoption of UML has progressed slowly [71]. High level of complexity of UML makes learning and adopting UML problematic, especially when IT people were lacking of the prerequisite skills [24]. In practice, IT professionals often draw diagrams with the symbols provided by the UML tool, but without the meanings those symbols are intended to provide. Dzidek et al. [27] argued that there is little reported evaluation of the adoption of UML. Moreover, to date, most of the studies on UML are technically oriented [6], and there is little empirical research on UML adoption reported. Using UML adoption as a vehicle to study technology adoption in organizations will shed light on better understanding the adoption of this important technology in an organizational setting. Our results suggest that technology characteristics, organization technology, and organization environment strongly affect UML adoption. In addition, larger organizations and organizations with a higher level of process maturity and strong presence of process champion are more likely to adopt UML. However, we find that some of the dimensions of both organizational and project cultures positively affect UML adoption while other dimensions of cultures have no direct impact on UML adoption. This study makes an important theoretical contribution to IT adoption literature by being the first to construct and test a comprehensive and integrated model integrating both organizational culture theory and the TOE framework. Move over, we argue that the proposed research framework for UML adoption is developed based on matching type of innovation for UML (hybrid innovation type) with the viable diffusion approach. Our findings also suggest that our holistic UML adoption model is superior to the traditional TOE framework in predictive power. In addition, this study also adds to the literature on UML adoption across varied company sizes and in different industries. Despite its numerous perceived benefits, UML has met with relatively slow acceptance [71]. This paper, therefore, provides important managerial implications for both developers and managers in better understanding the driving forces of adopting UML, and thus implementing such applications more efficiently. The paper will proceed as follows. First, we review relevant literature. We then develop our hypotheses. After developing our hypotheses, we describe our survey methodology, our multi-sector sample (N = 251), and our regression analyses. We then present our results and discuss the implications of our findings for both researchers and practitioners. Finally, we restate and summarize our contributions.

Multinomial logistic regression was employed in this study as the analysis allowed us to assess the effects of fourteen variables on UML adoption (Table 7). The R-Squares for the regressions for both holistic model and TOE model are 0.38 (F = 51.37) and 0.22 (F = 19.47) respectively; thus, the proposed predictors have satisfactory explanatory power.Hypothesis 1a suggests that a firm's higher level of perceived benefits of UML has a positive effect on the firm's adoption of UML. Table 7 shows perceived benefits of UML has significant influence on adoption (β = .42, р < .001), thus supporting Hypothesis 1a. By the same token, Hypotheses 2a (β = .39, р < .001), 3 (β = .26, р < .01), 4 (β = .20, р < .05), 5a (β = .34, р < .001), 5b (β = .27, р < .01), 6b (β = .23, р < .05), 6c (β = .28, р < .01), 7a (β = .22, р < .05), and 7b (β = .21, р < .05) are all supported correspondingly. Hypothesis 1b suggests that UML complexity has an adversary impact on UML adoption, while Hypothesis 2b suggests that satisfaction with the existing SAD tool might also diminish the willingness of adoption of UML. According to Table 7, UML complexity and satisfaction of the existing SAD tool are both negatively related to UML adoption (i.e., β = −.26, р < .01 and β = −.28, р < .01 respectively) in support of Hypothesis 1b and Hypothesis 2b. Table 7 indicated that our hypotheses, i.e., H6a (β = .07) and H7c (β = .05) were not supported. Our results also show that different industries (β = .02, .05, .03, .03, and .04 for respective industry) have no significant effects on UML adoption in the regression model. Table 7 also shows that the holistic model with adjusted R-Square value of 0.38 outperforms the TOE model (adjusted R-square value of 0.22) in terms of predictability. We also performed a simple t-test and the result (t = 2.17) indicates that the adjusted R-Square of the holistic model is significantly better than that of the TOE model.