316 zyxwvutsrqponmlkjihg IEEE TRANSACTIONS ON ENGINEERING MANAGEMENT, VOL. zyxwvu 38, NO. zyxwv 4, NOVEMBER 1991 Successful System Development: The Effect of Situational Factors on Alternate User Roles Stephen R. Hawk and Brian L. Dos Santos Abstract-It is widely believed among information systems professionals that users should be involved during systems devel- opment and, numerous studies have been undertaken to deter- mine the impact of user involvement. The results have been mixed, with only some studies concluding that user involvement is beneficial. This paper reports on the results of a field study that sought to determine the impact of two forms of user involvement, namely, (1) user participation, and (2) user leader- ship. In addition, this study sought to determine the impact of two situational factors on the impact of user participation and user leadership. Neither user participation nor user leadership were found to be related to user information satisfaction zyxwvutsrq (L’ZS). Both forms of user involvement however, were found to have a more positive relationship with UZS under certain circum- stances. User participation was more positively related to UIS when use of the system was for decision support and when users were at higher levels in the organization. User leadership was more positively related to UIS when users were at higher levels in the organization. INTRODUCTION VER the past three decades, investments in computer- 0 based information systems (IS) have increased substan- tially in most organizations. Yet, even today, many IS devel- opment efforts may be deemed failures. Among the many solutions prescribed to remedy the problem, perhaps the best known and most widely investigated is “user involvement” in the systems development process. Researchers and IS professionals generally agree that IS success can be enhanced by involving users in the development process [33], [52], [53]. A role frequently prescribed for users is participation in the analysis and design activities undertaken by the project team [ l l ] , [37]. Participation includes working with IS pro- fessionals in determining system requirements, developing and evaluating system design, and planning system imple- mentation. It has also been suggested that user involvement should go beyond mere participation, to include leadership of the devel- opment team [13], [25], [73]. The user’s role in this case expands to include planning, coordinating, and controlling the activities of those involved in the development effort. Although many researchers have investigated the benefits of user participation in doing analysis and design work, few have examined the impact of user leadership [33]. Manuscript received April 12, 1990; revised October 1, 1990. The review S. R. Hawk is with the Stuart School of Business, lllinois Institute of B. L. Dos Santos is with the Krannert Graduate School of Management, IEEE Log Number 9102973. of this paper was processed by E. Geisler. Technology, Chicago, IL 60616. Purdue University, West Lafayette, IN 47907. A number of IS authors have suggested that the benefits of user involvement are contingent upon situational factors spe- cific to the system being developed 1241, [33], 1601. Little effort, however, has been made to assess the impact of situational factors on the benefits of either form of user involvement [33]. Because both user participation and user leadership demand considerable time and effort from users, it is important to assess the benefits of both approaches, and to determine whether situational factors affect their impact on development outcomes. This research study sought to determine the impact of user involvement (i.e., participation and leadership) on user satis- faction with the resulting system. It also sought to determine the effect of user level and system type on the impact of user involvement. In the next section, the hypotheses used in this study are developed. Ensuing sections describe and discuss the research methods, results, and conclusions. RESEARCH MODEL AND HYPOTHESES It is widely accepted that user involvement in the develop- ment of computer-based information systems (IS) can posi- tively affect their successful introduction into the organiza- tion. Despite the wide acceptance of this view, the benefits of user involvement have not been adequately confirmed by prior research. In a review of user involvement research, Ives and Olson [33] reported that although some studies found a positive relationship between user involvement and system success, many others produced mixed, nonsignificant, or even negative results. Recent research, however, has been more supportive of the benefits of user involvement [6], [7], As indicated earlier, both user participation in and user leadership of the development process are recommended in the IS literature. Further, the decision-making literature sug- gests that the impact of user involvement may differ depend- ing upon whether users take on a participative versus a leadership role. Yet, little effort has been made to study the impact of these different forms of involvement. Hence, the impact of two user roles were investigated: (1) user participa- tion in the development effort, and (2) user leadership of the development effort. The IS literature suggests that situational factors moderate the impact of user involvement on system success [19], [33], [MI, [60], [MI. Participative decision making (PDM) re- search provides evidence that the impacts of different forms of participation and leadership are contingent upon the situa- tion, supporting a similar claim concerning the contingent nature of user involvement’s impact [41], [67]. Literature [191, ~ 4 1 . 0018-9391,’91$01.00 01991 lEEE