Effective pair programming relies heavily on optimal partner compatibility, a requirement that is often difficult to scale manually in software engineering education. This study presents the empirical validation of Codenote, an AI-driven Integrated Development Environment (IDE) designed to automate personality-aware group formation. By integrating a behavioral analysis mechanism, Codenote infers student personality traits from coding patterns to construct complementary pairs, thereby facilitating intelligent collaborative learning. To validate the system’s effectiveness, a controlled experiment was conducted to assess the impact of this AI-mediated pairing strategy on students’ self-efficacy across adaptive, innovative, and persuasive domains. Results indicate that students paired via Codenote’s complementary algorithm demonstrated significantly higher adaptive self-efficacy compared to the control group. This suggests that the system’s ability to expose learners to diverse problem-solving perspectives effectively enhances their confidence in managing complex and dynamic programming tasks. While no significant improvements were observed in innovative or persuasive self-efficacy, these findings identify specific directions for future system iterations, such as integrating automated scaffolding for creative exploration and leadership. Overall, this study demonstrates the viability of Codenote as an intelligent tool for scaling personalized instruction and highlights the crucial role of AI-driven pairing strategies in fostering psychological readiness for collaborative problem solving.
Lin et al. (Thu,) studied this question.