A Replication Study: Where and When Should Defects be Re-Assigned
Tamer Abdou, Ayşe Bener · 2016
Software development organizations often need to balance productivity and sustainable profitability. To keep a balance, attention needs to be given as to where and when defects are re-assigned. This paper reports a replicated experiment comparing the prediction differences of re-assigned defects between proprietary and open-source software (OSS) projects. We pursue a quantitative approach based on logistic regression that relies on two OSS defect tracking datasets: namely, Mozilla and Eclipse. Here, we focus our attention to the reputation of software developers and where it affects the re-assigned defects. We explore the correlation between reassigned defects and other related factors such as the reputation of the software developer. The replication of the original quantitative approach aims to verify the results of the original experiment on OSS defect datasets, rather than on proprietary ones.