Measuring students' source code quality in software development projects through commit-impact analysis

This repository contains the data and figures used for the paper "Measuring students' source code quality in software development projects through commit-impact analysis".

Click to expand the reference to the paper

@inproceedings{hamer2021measuring, 
  title={Measuring students' source code quality in software development projects through commit-impact analysis},
  author={Hamer, Sivana and Quesada-L{\'o}pez, Christian and Mart{\'i}nez, Alexandra and Jenkins, Marcelo},
  booktitle={International Conference on Information Technology \& Systems},
  year={2021},
  organization={Springer}
}

Abstract

Evaluating the quality of software engineering projects in university courses is challenging because it evolves over time and is a time consuming task. Students applying software quality principles need early and constant feedback on their projects to improve their technical competence as software developers. We conducted a case study to explore whether student' changes have an impact on the project quality by mining a Git repository. We analyzed a total of 2253 changes (commits) from an undergraduate software engineering project to understand the impact on quality of each change measuring metrics (complexity, duplication, maintainability, and security) mined from the repository. This analysis allowed us to identify from students' contributions challenges and improvement opportunities in engineering practices. As future work, we plan to analyze more projects and extend our analysis with more software metrics.

Index terms

software engineering education, mining software repositories, commit-impact analysis, software quality, project based learning

Data and figures

All the data and figures can be found in the REPORT.

License

This work is licensed under a Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International License