MAADS: Mixed-Methods Approach for the Analysis of Debugging Sequences of Beginner Programmers

Jemmali, C., Kleinman, E., Bunian, S., Almeda, M. V., Rowe, E., & El-Nasr, M. S. (2020, March). MAADS: Mixed-Methods Approach for the Analysis of Debugging Sequences of Beginner Programmers. In Proceedings of the 52nd ACM SIGCSE Technical Symposium on Computer Science Education.

Keywords

Debugging, Programming, Process, Beginners, Methodology

Abstract

Debugging is a cornerstone of programming and has been shown to be especially problematic for beginners. While there has been some work trying to understand the difficulties that beginners face with debugging, investigating common mistakes or specific error types they struggle with, there is little work that focuses on in-depth analysis of how novice programmers approach debugging, and how it changes over time. In this paper, we present MAADS (MixedMethods Approach for the Analysis of Debugging Sequences), a scalable and generalizable approach that combines quantitative and qualitative methods by using a state/action representation and visualization to gain knowledge about the debugging process through a step by step analysis. To demonstrate the utility of MAADS, we used it to analyze the debugging processes of middle school students who developed code within May’s Journey, a game designed to teach basic programming principles. The approach showed great utility in identifying differences in students’ debugging techniques and learning paths.