This study proposes a Debugging Method Suggestion System (DMSS) using past error information to address the difficulties faced by learners when solving bugs in programming-learning environments. The DMSS automatically collects error information generated when students work on assignments, allowing instructors to add annotations describing the causes and solutions for these errors. When other students using the same programming materials encounter similar errors, the system presents them with the accumulated similar error information and advice. This system was implemented to assist students in developing a learning web application using the Ruby on Rails framework. Its effectiveness was verified by an experiment involving 53 university students. The results showed that the debugging success rate when using the DMSS for unknown errors was more than twice that of cases where DMSS was not used, with particularly significant support effects for first-time errors. However, differences in effectiveness based on error difficulty and prior knowledge were observed, suggesting that content and quality of advice are more important for complex errors.
Keiichi Takahashi (Wed,) studied this question.