Teaching evidence-based software engineering

Abstract

In this paper, we share our experiences about teaching evidence-based software engineering to students of a Master degree program in Computer Science. We provided a semester-long course, composed of lessons about empirical and experimental methods. It also included a collaborative project concerning a systematic mapping study of the challenges in the adoption of open source software in a business context. All students collaborated on the project by analyzing emerging results in the scientific literature. They evaluated the proposals in terms of level of novelty and evidence and delivered a complete report, which summarized the risk factors in the adoption of open source software and offers technical knowledge about evolutionary patterns and development community support, with practical implications. As a side effect, this problem-based learning approach provides a positive impact in terms of students' participation, teamwork attitude, professional interest in open source software, and exam passing.