Software Quality Alignment : Evaluation and Understanding

Abstract: Background: The software development environment is growing increasingly complex, with a greater diversity of stakeholders involved in product development. Moves towards global software development with onshoring, offshoring, insourcing and outsourcing have seen a range of stakeholders introduced to the software development process, each with their own incentives and understanding of their product. These differences between the stakeholders can be especially problematic with regard to aspects of software quality. The aspects are often not clearly and explicitly defined for a product, but still essential for its long-term sustainability. Research shows that software projects are more likely to succeed when the stakeholders share a common understanding of software quality. Objectives: This thesis has two main objectives. The first is to develop a method to determine the level of alignment between stakeholders with regard to the priority given to aspects of software quality. Given the ability to understand the levels of alignment between stakeholders, the second objective is to identify factors that support and impair this alignment. Both the method and the identified factors will help software development organisations create work environments that are better able to foster a common set of priorities with respect to software quality. Method: The primary research method employed throughout this thesis is case study research. In total, six case studies are presented, all conducted in large or multinational companies. A range of data collection techniques have been used, including questionnaires, semi-structured interviews and workshops. Results: A method to determine the level of alignment between stakeholders on the priority given to aspects of software quality is presented—the Stakeholder Alignment Assessment Method for Software Quality (SAAM-SQ). It is developed by drawing upon a systematic literature review and the experience of conducting a related case study. The method is then refined and extended through the experience gained from its repeated application in a series of case studies. These case studies are further used to identify factors that support and impair alignment in a range of different software development contexts. The contexts studied include onshore insourcing, onshore outsourcing, offshore insourcing and offshore outsourcing. Conclusion: SAAM-SQ is found to be robust, being successfully applied to case studies covering a range of different software development contexts. The factors identified from the case studies as supporting or impairing alignment confirm and extend research in the global software development domain.