Extreme programming: a survey of empirical data from a controlled case study

Extreme programming (XP) is a well known agile software development method. While a number of experience reports have been published in recent years, agile software development in general and XP in particular have strongly been criticized for the lack of empirical data. This paper reports a survey of the empirical data obtained from a controlled case study on extreme programming in practical settings. Thus, no hypotheses were set a priori. Four software engineers were acquired to implement a Web-based system (7698 Locs, 820 hours) for data management in a delivery schedule of eight weeks. Development environment was close to the agile home ground. Collected empirical data is grounded on three basic data points: time, size and defects. Data is organized around five system releases, each which were tested by 17 customer testers. System release defect-density was 1.43 defects/KLOC, team overall productivity 16.90 Locs/hour and rework costs were 9.8% of the total development effort. The implications of this study are discussed.