The main goal of this study was to evaluate the impact of ChatGPT on student performance in a project-based learning environment, specifically through a software development competition.
Methods: The study involved a four-week software development competition with 36 students, structured in two rounds. Participants used ChatGPT during various stages of their projects: idea planning, documentation, coding, debugging, and quality assurance. The performance was assessed based on metrics such as code quality, innovation, and adherence to project requirements. Surveys and usage logs were analyzed to measure the tool's impact.
Key Findings:
- Students using ChatGPT extensively had higher project completion rates and better scores.
- Top performers in the competition integrated ChatGPT more thoroughly across different stages of their projects.
- Experience with large language models (LLMs) like GPT contributed to better performance.
- High levels of student satisfaction with ChatGPT were reported, highlighting its benefits in enhancing learning experiences and project outcomes.
Implications: The findings suggest that integrating ChatGPT into educational settings can significantly enhance project-based learning by improving student performance and satisfaction. It supports the use of generative AI tools to enrich the learning process and develop practical skills.
Limitations: The study's limitations include a small sample size and a focus on a specific educational context (software development competition). Potential biases may arise from the self-reported nature of surveys and the competitive environment.
Future Directions: Future research should explore the long-term impact of ChatGPT in diverse educational settings and involve larger participant groups. Additional studies could focus on refining AI tools to support a broader range of tasks and further investigate their role in specialized areas like security and quality assurance.
Title and Authors: "ChatGPT and Its Educational Impact: Insights from a Software Development Competition" by Sunhee Hwang, Yudoo Kim, and Heejin Lee.
Published On: August 26, 2024
Published By: KDD AI4EDU Workshop '24, ACM