ASP 사용자 만족도에 영향을 미치는 변수에 대한 탐색적 연구

Abstract

In the perspective of small and mid-sized companies, ASP is a new business model that allows them to implement efficiently in short time at low initial costs. Due to this usefulness, Gartner Group predicted that ASP industry would grow into 253 million dollar market and would generate 100% growth rate annually (Cole et., 1999). Also, IDC predicted the growth of industrial APS industry to 20 million dollar market up to year of 2003 (IDC, 1999). But in the case of Korean market, ASP is facing difficulty in expansion of market size due to poor recognition of ASP service, and lack of credibility and failure developing ASP business model to fit various industry. Also at this limited market, as more vendors enter the outsourcing business and organizations battle to remain competitive and technologically competent, service quality and system quality are of fundamental importance. (Grover, Cheon, Teng; 1996). Therefore, to remain competitive, determining which service quality factors and system quality factors affect user satisfaction is important. Through this research, the purpose of this research is to determine which system and service quality factors affect user company satisfaction. This research will review previous researches on system quality and service quality factors and provide concise review of the previous researches. Also this research will exam whether the IS level of users knowledge and involvement is considered important and needed in providing general or specific ASP. Through determining this, companies will be able to decide which factors among system quality and service quality to concentrate on to improve user satisfaction. This research researched system quality factors and service quality factors from previous researches such as Grover, Cheon & Teng (1996) which affects information system user satisfaction. Mediator factors which are level of IS knowledge and user involvement is also reviewed. Since ASP compromised of application, network, and data and these factors should inter-coordinate well in order to achieve successful ASP. The system quality that we want to measure through this research is the ability to process provided the applications such as software from ASP vendor and IT infra structure, network connectivity, users PC is included in the system. In IS area, previous studies were done about system quality. Among them, the constructs that fit into ASP context are accessibility, currency, reliability, security, and flexibility. Pitt et. al argues that service quality needs to be considered as an additional measure of IS success (Pitt et., 1995). ASP users do not just want a PC and ASP application. Rather, they seek the system that satisfies ASP needs. Thus, the service quality should be considered and as factors for service quality, reliability, assurance, responsiveness has been chosen in this research. According to Joshi (1990), level of user's IS knowledge is considered as one of the major factor that affect user satisfaction. In MIS context, level of user's IS knowledge refers to issues such as training and existing knowledge of information system (Swanson, 1947). In ASP context, level of users knowledge may or may not be considered important due to characteristic of ASP. Therefore, IS knowledge should be tested as a moderator. According to Joshi (1990), users involvement is considered as one of the major factor that affect user satisfaction. User involvement in the process of developing information systems has been known to be a critical component of eventual success. (Jiang, Chen, and Klein, 2002) In MIS context, user's involvement refers to issues such as involvement in system design and system implementation. (Swanson, 1947). In ASP context, users involvement is also considered important in a way that using ASP application is part of business process reengineering. For successful implementation of ASP application use, users involvement should be carefully reviewed. While user satisfaction is considered a dependent variable in this study, system quality and service quality are considered as independent variables that cause the user satisfaction. As a moderator, this study also investigates whether level of user's IS knowledge and Users Involvement affect user satisfaction. Expert interviews and survey instrument were used in this study. The survey was given to managers who were in charge of ASP or made contract with ASP vendor. Of the 80 survey instruments distributed, 52 were returned and usable (65 percent response rate). To test the relationship- (1) between dependent variable and each of independent variables and (2) between each pair of independent variables Pearson correlation coefficients was used. After the test, there seem to have relationship between system quality, service quality and user satisfaction. The validity and reliability of the instrument were evaluated. Construct validity was examined by factor analysis, using the principal components method with a varimax rotation. After factor analysis, six factors were named according to their characteristics. Credibility, System Usability, System support, Security, System Currency, Ubiquitousness. Six factors resulted from factor analysis is used to examine whether they affect user satisfaction and moderator from the original research model is added to develop following hypotheses. As a result, a regression model, five factors which are Credibility, System Usability, System Support, Security, System Currency was proven to be significant (p=0.000, p=0.000, p=0.000, p=0.023, p=0.024)) and Ubiquitousness was proven to be insignificant. Therefore, hypotheses HI, H2, H3, H4, and H5 were accepted and H6 is not accepted. The most important factor that can explain user satisfaction is Credibility (Beta=0.485), System Currency (Beta=.427) Security (Beta=0.344), System Usability (Beta=.212), and System Support (Beta=.210) listed in order of most the important to the least. The result shows that the moderating effect (Level of IT knowledge) was statistically not significant on H1a, H2a, H3a, H5a but significant on H4a. Overall, Level of IT knowledge had no moderating effect on user satisfaction. Therefore, hypotheses H1a, H2a, H3a, H5a were not supportive and H4a were supportive. Model 2 in Table 9 shows that the moderating effect (User's Involvement) was statistically significant on H1b, H2b, H3b, H4b, and H5b. Overall, Users Involvement has moderating effect on user satisfaction. After careful interview, the reason for moderating effect was although easiness of use of general software, modification was required and therefore, users involvement needed.