Teaching Empirical Software Engineering Using Expert Teams

Empirical software engineering aims at making software engineering claims measurable, i.e., to analyze and understand phenomena in software engineering and to evaluate software engineering approaches and solutions. Due to the involvement of humans and the multitude of fields for which software is crucial, software engineering is considered hard to teach. Yet, empirical software engineering increases this difficulty by adding the scientific method as extra dimension. In this paper, we present a Master-level course on empirical software engineering in which different empirical instruments are utilized to carry out mini-projects, i.e., students learn about scientific work by doing scientific work. To manage the high number of about 70 students enrolled in this course, a seminar-like learning model is used in which students form expert teams. Beyond the base knowledge, expert teams obtain an extra specific expertise that they offer as service to other teams, thus, fostering cross-team collaboration. The paper outlines the general course setup, topics addressed, and it provides initial lessons learned.