Stimulating Critical Thinking in a Web Programming Module with Generative AI Tools

Abstract

Web frameworks have significantly changed how developers create web applications for the Internet. Thanks to pre-defined libraries, these frameworks not only accelerate development time but also reduce the amount of code developers need to write. However, to get the most out of the frameworks and libraries, developers need to have a deep understanding of core web programming languages. This allows them to write efficient code, troubleshoot effectively, and push the boundaries of what the frameworks can achieve. The same principle applies to Generative Artificial Intelligence (AI) tools, as they have the potential to enhance a developer's toolkit. However, they will only be useful if the developer has sound fundamental knowledge to verify the output from these tools. Educators in higher education face a similar predicament with the widespread use of Generative AI tools by learners. Many learners rely on these tools as a go-to solution without being able to verify or fully comprehend the output, leading to shallow understanding. The work in this paper outlines an approach used in a first-year web programming module within the School of Electronic Engineering and Computer Science at Queen Mary University of London, where learners were encouraged to use Generative AI tools to stimulate critical thinking when conducting assessments. Specifically, GitHub CoPilot was used as a pair programmer, and ChatGPT served as a peer reviewer. In this context, the peer reviewer's role was to help the learner reflect on the tool's output. The aim of this study was to explore the design of active learning activities that incorporate Generative AI tools for web programming to foster critical thinking practices among learners. To evaluate our approach, we employed a critical thinking self-evaluation questionnaire instrument, where learners' opinions and customs were surveyed before and after both of the assignments.