As undergraduate computer science classes grow, instructor workload also increases. At scale, it is hard to know which students need extra help–much less inspire, challenge, and connect with learners. But there is hope; with more data, algorithms and systems can more accurately identify student outliers and visualize student trends. In this talk I discuss an approach to characterizing student work in a particular CS1 graphics-based programming assignment by connecting compiled image output to functional milestones. I also introduce two systems for understanding student progress that are used in classrooms today: TMOSS detects excessive collaboration during a programming assignment, while Pensieve facilitates teacher-student conversations on student metacognition. My research aims to give teachers the critical information they need to directly support students even in the largest of classrooms.
Lisa Yan is a PhD candidate in Electrical Engineering at Stanford University. Her research leverages in-depth student data in systems that encourage student-teacher interaction in large introductory computer science courses. Lisa holds an MS in Electrical Engineering from Stanford University and a BS in Electrical Engineering and Computer Science from UC Berkeley; she is an NSF Graduate Research Fellow.