Assisting students with typical programming errors during a coding session

Abstract

Typical programming errors that introductory students make are known to stall and frustrate them during their first few lab coding sessions. Traditionally, lab instructors help with the process of correcting these errors; however, with medium or large size computer labs, the degree of interaction between students and the instructor tends to decline; leaving some students feeling unattended and subsequently perhaps uninterested. To help automate the process of finding and correcting these errors we have devised a solution that collects periodical or on-compile time code snapshots which we analyze upon unsuccessful compilation or whenever a lengthy stall is detected. The analysis is then used to provide feedback to students directly into their Integrated Development Environment (IDE), and generate useful reports that can be perused by both instructors and students later on. This article describes the design and implementation of our solution in the context of BlueJ; a common first year IDE.