Application software development in the business world traditionally centered around business applications such as accounting, production, and marketing, where users and managers understood the application area. Today, software development has transitioned to a position where business need is a wicked problem that requires abductive reasoning (that is, the logic of what might be). In the old days, the business requirements were understood and one of the main software development challenges was determining how to extract the software requirements from the business understanding embedded within the user community and then building modifiable systems. Today, to address the business need, business users and software engineers need to collaborate and concurrently shape both business understanding and software requirements to build evolving systems. This has given impetus to the agile development movement, where the client is an integral part of the software development process, and requirements are discovered and formulated based on successive evolutions of the software. An open question is: Are agile development methodologies more effective than traditional life cycle approaches for today’s needs?

This paper is a small step toward developing a comprehensive answer to this question. There is no guarantee that all agile methodologies are equally effective and we cannot be certain that all agile teams adhere to the tenets of agile methodology. For those reasons, an important first step in agile methodology theory development is to establish whether fidelity to agile methodology is significant in achieving team performance. Specifically, this paper explores whether adherence to the tenets of the extreme programming (XP) methodology is essential for the success of XP-based software development projects. In addition, because team factors, such as trust, cohesiveness, and cooperation, can impact software team performance, the XP methodology fidelity research model presented in this paper explicitly incorporates both XP-specific team factors and general team factors. The empirical findings of the study are: (1) that XP practices and XP-specific team factors are significant for team performance, and (2) that only one general team factor, namely cooperation, is significant for team performance.

Because cooperation is a necessary ingredient for “XP-specific team factor [practices such as] continuous integration, coding standards, and collective code ownership,” the two empirical findings suggest that “teams that stick to the core XP practices [can] gain a competitive advantage ... over teams that are cohesive and work well together.”

The main weakness of the paper is the manner in which the independent variable, team performance, is operationalized and measured. This weakness renders the findings of the paper questionable. However, the development of the research model is an important contribution of this paper to the literature.