workflows like Continuous Delivery or asynchronous code reviews
are added to the syllabus [12]. The instructor has to
introduce these workflows to students with different backgrounds
in software engineering, and then continuously ensure
that the students are able to successfully apply the
learned concepts and techniques in their projects. This can
be very time-consuming, which is why the instructor needs
a way to quickly assess each project’s status and identify
potential problems which call for action.
We propose the usage of metrics to increase the manageability
of such a software engineering project course. We
regularly run a capstone course with over 100 students developing
applications in 10-12 parallel projects with customers
from industry over the course of three months [1]. During
the summer term of 2015, we selected and evaluated metrics
for the key workflows of our course. In this paper we
present our approach and explain the benefits and challenges
of using metrics to evaluate the progress of student projects.
We start with an overview of the project course and then
establish goals for selecting a set of metrics. We then describe
our workflows and discuss how we selected metrics to
measure their adaptation and usage. Our focus hereby is on
three key workflows, namely Merge Management, Continuous
Integration and Continuous Delivery. We describe our
experiences with using these metrics for the management
of our project course in a case study we conducted in the
summer term of 2015. Finally, we discuss the results and
provide an outlook on the next steps we want to take.