Current Status of AI Programming Education for Non-Majors

This study conducted a meta-analysis to analyze the effectiveness of artificial intelligence education targeted at non-majors. Among 20 papers published until January 2024 on AI programming education for non-majors, 12 papers including statistical data were selected, excluding 4 papers that were simple surveys or singlegroup studies. The final selection consisted of 8 papers, all of which were published in academic journals. The research findings showed that the overall effect size of AI programming education was 0.502, indicating a moderate level of effectiveness and substantiating its practical educational impact significantly with a significance level of <.001. Analyzing the educational effects based on programming language types, block-based educational programming languages demonstrated the highest average effect size of 0.713. This suggests that educational programming languages provide learners with opportunities to experience AI without significant difficulty, thereby fostering active participation and interest in classes. Adjusting the difficulty and accessibility of educational content is essential to facilitate non-majors’ easier access to AI and programming. This integration of domain knowledge and AI technology can lead to solving new problems and generating innovative ideas, thereby promoting the application of AI in various academic fields. This study is expected to contribute to the effective design of AI programming education for non-majors and provide direction for future research and educational program development.