Capturing Student Feedback and Emotions in Large Computing Courses: A Sentiment Analysis Approach

Enrollment numbers in computer science courses are higher than ever and keep growing. This renders communication and interaction of instructors with individual students extremely challenging, leading to an increase in anonymity (anonymity gap). Especially when students struggle in computing courses, personalized help is crucial for them to overcome their problems and frustration and eventually succeed in their studies. At the same time detecting students' misconceptions and gathering feedback at scale is time consuming, resulting in a lack of unbiased feedback available to course instructors (feedback gap). Real-time student feedback is a crucial source for instructors to adapt their teaching pace, teaching materials, or course content during the course of the semester to cater to an increasingly diverse student population. In this paper, we investigate a scalable approach to collect and analyze student feedback and emotions. We find that sentiment analysis can efficiently capture student emotions, bearing the potential to lessen both the anonymity and feedback gaps.