City Research Online

Abstract

In this thesis I analyse how QA and measurement programmes are implemented in the software industry. My thesis is that, contrary to popular belief, how companies implement QA and measurement is as important to the ultimate success of a programme as what QA and measurement tools and techniques are implemented.

I have combined input from over 300 software developers and managers in 26 different companies, with the analysis of all the relevant publicly available information on companies’ experiences to produce a definitive study into implementing QA and measurement in the software industry. In this study I:

• identify those QA and measurement practices considered to be state-of-the-art;
• quantify the penetration of state-of-the-art practices;
• report on the state-of-the-practice;
• assess the effectiveness of the state-of-the-practice;
• present guidelines for improved practice.

I tested a variety of hypotheses about implementing QA and measurement programmes. My key findings about these hypotheses include:

• Some companies have a large gap between what is officially said to be implemented and what is actually implemented.
• Practitioners had a significant impact on the success of programmes. Furthermore, I confirmed my hypotheses about managers and developers having different attitudes to QA and measurement. However, I was unable to find convincing evidence to confirm that the demographics of practitioners affected their attitudes.
• Many of my hypotheses about the implementation process were confirmed. In particular I found that doing background research was related to success, but that very few companies did research. However, I was unable to find convincing evidence to confirm the importance of feedback to practitioners.

Overall my study reveals that the quality of the implementation process across the software industry is, in practice, very poor. Indeed, I show that the gap between state-of-the-art and state-of-the practice is large. I conclude that until the software industry puts as much effort into developing strategies for implementing QA and measurement as it does into developing tools and techniques, then effective and optimised programmes will remain rare.

Publication Type:	Thesis (Doctoral)
Subjects:	Q Science > QA Mathematics > QA76 Computer software
Departments:	School of Science & Technology > Computer Science > Software Reliability School of Science & Technology > School of Science & Technology Doctoral Theses Doctoral Theses

Export

Downloads