The purpose of this study is to investigate the value of contingency factors on post-adoptive beliefs that influence a user’s satisfaction and intention in the context of mandatory continued IS usage. It is certain that long-term viability of an IS and its eventual success depend on its continued use rather than its first-time use(Bhattacherjee, 2001b). If management wants to promote continued IS usage behavior of users, it is necessary to understand what drives it. We are great on the two existing circumstances. First, when does the IS usage behavior really begin? When a user does decide to accept it? Or when the user uses it to carry out assigned work tasks? Because of the path-dependent nature of IT adoption and use processes in general(Rogers, 2003), post-adoptive behavior must be framed within this larger context. However, we only focus on post-adoptive behavior, because obvious distinctions have been observed between pre-adoptive and post-adoptive beliefs and behaviors(Karahanna et al.,1999; Oliver, 1980). TAM has primarily focused on adoption under the implicit assumption that users' attitude and intention are mainly determined by pre-adoptive cognitive beliefs. It does not reveal what is really driving continued usage behavior above and beyond the antecedents derived from TAM researches. Alternatively, the expectation-confirmation model of IS continuous (ECM-IS continuous)(by Bhattacherjee, 2001b) based on a solid theoretical foundation that focuses on an individual’s psychological motives during post-adoption, and it has been successfully adapted to the IS continuous context. As mentioned reasons, we decided to use Bhattacherjee’s model instead of TAM as the departure of our study. Second, this study focuses on the environments where usage is mandatory. In today’s organizations, tasks are increasingly depending on work systems, forcing their members to use the systems to increase their job productivity. The contingency approach suggests that a number of contextual variables influence the performance of IS; the better the 'fit' between these variables and the design and use of the IS, the better the IS performance(Weill and Olson, 1989). Previous researches have in particular emphasized the task-technology fit, job relevance, or compatibility with work in the context of IS mandatory(Kim et al., 2007). Task characteristics have a huge effect on a user’s satisfaction on their IS and success of the system (Raymond and Bergeron, 1992; Sanders and Courtney, 1985). In addition, in IS field it has been identified that individual factors can be as important as task factors for the success of IS. User characteristics are also found to affect the post-adoptive beliefs. We consider users’individual characteristic as well as work task characteristic as contingent factors in the context of mandatory IS usage. Accordingly, this paper proposes a contingency approach relating selected task characteristics and user characteristics on ECM-IS continuous in mandated environments. In this research,task characteristics were include in task variability and analyzability, and user characteristics were include system-specific self-efficacy and user education and training, which affects to form users’ cognitive belief such as expectation-confirmation and perceived usefulness during post-adoption. This paper attempts to expand the set of contingency factors in the ECM-IS continuous, in order to confirm the application of the ECM-IS continuous in IS mandatory usage context. The expanded ECM-IS continuous, incorporating the contingency factors of task variability,task analyzability, system-specific self-efficacy, and user education and training, was empirically validated with data collected from a field survey of 169 existing users of e-customs system. We used partial least square(PLS) to examine the reliability and validity of the measures,and to analyze this extended ECM-IS continuous. We chose PLS for two reasons. First, our model has formative constructs. PLS uses components-based algorithm and can estimate formative constructs. Second, PLS is more appropriate when the research model is in an early stage of development. The data analysis showed that the expanded ECM-IS continuous has good explanatory power(R2=27.5% of IS continuance intention, R2=28.7% of satisfaction, R2=30.5% of perceived usefulness, and R2=14.5% of confirmation,), with six out of nine paths supported. We conclude that the extended ECM-IS continuous for mandatory usage shows a reasonably good fit. The test of model was executed in the order of two research questions. First, task variability of task characteristics had the direct effects on confirmation(β=.247, p <.001) and perceived usefulness(β=.249, p <.001). And task analyzability had only the direct effects on perceived usefulness(β=.175, p <.01) but statistically, no confirmation. In case of user characteristics,system-specific self-efficacy had not only the direct effects on confirmation but also on perceived usefulness. User education and training, however, had the direct effects on both confirmation (β=.306, p <.001) and perceived usefulness(β=.146, p <.05). Second, all path coefficients in the ECM-IS continuous have significant t-value. Specifically, perceived usefulness is an important construct in the ECM-IS continuous, where it has been shown to have a significant influence on continued IS intention(β=.366, p <.001), as well as on satisfaction (β=.480, p <.001) of that system. Also it is found in this study that confirmation is the most prominent variable influence on both satisfaction(β=.152, p <.05) and perceived usefulness(β=.130, p <.05). And finally, continuance intention had been affected by satisfaction(β=.232, p <.01). This study has two contributions. First, this is the first empirical study showing that contingency factors of task characteristics and user characteristics are the significant drivers of the post-adoptive beliefs of users. The significant effects of task variability and user education and training signify that the post-adoptive beliefs of confirmation and perceived usefulness can be depended on the contingency factors in organizational context. The researchers of ECM-IS continuous have hardly studied contingency factors, even though they have studied on the organizational systems. Second, this study can provide supplementary information that is relevant for understanding continued IS usage in a mandated use environment in an organization. Until recently IS continuance behavior has been studied from the individual viewpoint of a user, even though almost all the organization users are required to use a specific IS in order to perform their work tasks. Hence, the expanded ECM-IS continuous present managers and researchers with deeper insights into how to address IS users’ satisfaction and continued intention in mandated environment. Finally, our study compliments traditional studies of ECM-IS continuous. According to prior literature about of ECM-IS continuous, the constructs of confirmation, perceived usefulness,and satisfaction had effects on the IS continuance. This study proved that these constructs are still significant in IS mandatory usage environment.

강제적 사용 환경에서 정보시스템 지속에 대한 상황요인의 영향에 관한 연구