Software Quality Assurance in Agile and Waterfall Software Development Methodologies: a Gap Analysis

Abstract

The purpose of this study was to investigate whether the software development companies can achieve expected software quality through Agile development. In order to reach this goal, the first objective of the research was to identify the software quality factors through various quality models and quality management philosophies. Secondly, to identify the software development process models. Thirdly, to analyse the software quality difference between development methodologies in terms of selected quality factors. And finally to identify the development technique by which high quality software products could develop. The research was conducted in the Sri Lankan context focusing on software development companies registered with the Sri Lanka Exports Association. After the preliminary investigation on obtaining relevant information, four companies namely; Virtusa, Team Work, DMS and E- College were selected for the research. The second pilot survey reflected that it was impossible to collect data from clients. Thus, the research was aimed only on developer oriented quality factors. These selected factors include correctness, testability, changeability, install ability, time and budget. The research was confined to analyse the quality difference only between Waterfall and Agile since the second pilot survey revealed that the Waterfall is the most widely used in Sri Lanka. Both qualitative and quantitative research methodology was utilized in this study. The researcher tendered 190 questionnaires among testers, developers and QA leads via e-mails. The response rate for questionnaires was eighty one percent and the accepted rate was seventy two percent. Twelve interviews were carried out with the Project Managers to capture project related information. The result of the questionnaire revealed no significant difference between the two development methods in achieving correctness and install ability. Whereas, the difference in testability and changeability was significant and reflected that Agile is better than Waterfall. The cumulative analysis of the product quality factors showed that a high level of software quality can be achieved through Agile development. Analysis of the interviews reflected that there is no significant difference in the software project quality between the two development methods. The author recommends applying Agile techniques for software development projects where the requirements are complex, difficult to capture and frequently fluctuating (At situations where high degree of testing and changeability is required). Any method can be used if the main focus is to achieve only project quality.