Sun 14 Jun 2015 17:20 - 18:00 at C120-C121 - Session4

In MOOCs, a single programming exercise may produce thousands of solutions from learners. Understanding solution variation is important for providing appropriate feedback to students at scale. The wide variation among these solutions can be a source of pedagogically valuable examples, and can be used to refine the autograder for the exercise by exposing corner cases. We present OverCode, a system for visualizing and exploring thousands of programming solutions. OverCode uses both static and dynamic analysis to cluster similar solutions, and lets instructors further filter and cluster solutions based on different criteria. We evaluated OverCode against a non-clustering baseline in a within-subjects study with 24 teaching assistants, and found that the OverCode interface allows teachers to more quickly develop a high-level view of students’ understanding and misconceptions, and to provide feedback that is relevant to more students.

Sun 14 Jun

Displayed time zone: Tijuana, Baja California change

16:00 - 18:00
Session4PLOOC at C120-C121
16:00
40m
Talk
Personalized Mathematical Word Problem Generation
PLOOC
P: Alex Polozov University of Washington, Eleanor O'Rourke University of Washington, Adam Smith University of Washington, Luke Zettlemoyer University of Washington, Sumit Gulwani Microsoft Research, Zoran Popovic University of Washington
16:40
40m
Talk
Making Proof Tutors out of Proof Assistants
PLOOC
P: Peter-Michael Osera University of Pennsylvania, Steve Zdancewic
17:20
40m
Talk
OverCode: Visualizing Variation in Student Solutions to Programming Problems at Scale
PLOOC
Elena Glassman MIT, Jeremy Scott MIT, P: Rishabh Singh Microsoft Research, Philip Guo University of Rochester, Robert Miller MIT