A Methodology for the Categorisation of Software Projects in Nigeria Based on Performance

Abstract

Software engineering projects in Nigeria have been classified generally as a failure, challenged or successful with no proof that the projects fall into these categories. The main focus has been on cost and time overrun, and attention has not been given to check whether projects truly fall within the given categories. Discriminant analysis was employed to determine how the 30 selected projects in the public sector in Nigeria can be correctly classified. This study developed a method for determining the actual category of software engineering projects concerning the characteristics of projects as a failure, challenged, or successful. The developed model was used to reclassify the thirty (30) projects, and it was discovered that twenty- one (21) projects were correctly classified giving 70% of correctly classified projects while nine (9) were wrongly classified giving 30%. It is possible for projects to satisfy the established success criteria (requirements met on time and within budget) whereas the stigma of failure still exists in its result.