Designing a Conversational Agent as a Formative Course Evaluation Tool

Today’s graduating students face ever-changing environments when they enter their job life. Educational institutions must therefore continuously develop their course structure and content in order to prepare their students to be future employees. A very important means for developing the courses is the students’ course evaluations. Due to financial and organizational restrictions, these course evaluations are usually carried out quantitatively and at the end of the semester. However, past research has shown that this kind of evaluation faces certain constraints such as low acceptance rates, only time-related insights and low-quality answers that do not really help the lecturer to improve the course. Drawing on social response theory, we propose that conversational agents as a formative course evaluation tool are able to address the mentioned problems by interactively engaging with students. Therefore, we propose a set of design principles and evaluate them with our prototype Eva.