Improving systems design and implementation: An examination of the effect of good requirements definition on end-user satisfaction.

摘要

The failure of software development projects continues to be an issue in spite of all the technological tools available to assist the software developer. This study examines the Effect of Good Requirements Definition on End-user Satisfaction, and defines Good Requirements Definition as one that includes user participation. Since the best measure of success in the development of computer-based system appears to be the level of satisfaction the end-users have with the system, user involvement appears very important. This study sees the need for a combination of approaches that together make up Good Requirements Definition, and must include end-user involvement, communication, and management action from the very beginning of any development effort. The main research question examines the relationship between good requirements definition and end-user satisfaction. This was used to ultimately develop the investigative questions for the research. Hypotheses were developed for testing each area addressed by the questions. Data collection was done using self-administered survey questionnaires to look for responses on a Likert scale of 1 to 5. Nonparametric testing was used for analysis, and data were displayed using both bar and pie charts. Correlation was used to check for relationship between indicators of user participation and user satisfaction. The results show that (a) Better identification of user needs will include ensuring that a careful definition of the right set of requirements has been made. This can only happen when there is constant communications between users and developers. (b) The only way to ensure that the requirements definition process involves users is to communicate with users throughout the process. Management can facilitate this kind of open and lateral communication. However, the results also show that management has in general failed to demonstrate any commitment to open and lateral communications, and this apparent lack of management commitment to facilitate open communication could possibly exacerbate the very problems such as cost overruns, failed systems, and unusable systems that management would typically seek to avoid in any development process. (c) An examination of the main research question show that there is a positive linear relationship between indicators of user participation and user satisfaction.