The Impact of an AI-based Feedback System on the Improvement of Elementary Students’ Statistical Inquiry Question Posing: The Moderating Effects of AI Perception and Feedback Self-efficacy

Hunkoog Jho,Juyeong Lee
DOI: https://doi.org/10.31216/bdl.20240026
2024-09-30
Abstract:With the advent of the digital age, the amount of data has exponentially increased making statistical analysis essential for decision-making and problem-solving. This research examines the impact of an AI-based feedback system (FS) using the GPT-4 model on the ability of sixthgrade students to formulate statistical inquiry questions. Conducted at an elementary school in Gyeonggi-do, the research involved 95 students divided into experimental and control groups, who participated in an eight-session program. The experimental group received feedback from the AI-based FS, while the control group received teacher feedback. Pre- and post-tests measured the improvement in students’ statistical inquiry question levels. Additionally, the study analyzed how students’ self-efficacy regarding feedback and their perception of AI moderated the effectiveness of the FS. Results showed that the AI-based FS significantly improved the students’ ability to pose statistical inquiry questions compared to the control group. The study also found that a positive perception of AI enhanced the effectiveness of the FS, while self-efficacy regarding feedback did not show a significant impact. These findings suggest that AI-based FS can be an effective educational tool, particularly when students have a positive attitude toward AI. Future research should focus on developing fine-tuned AI-based FS capable of providing detailed feedback throughout all stages of statistical inquiry and investigate its effects on students with cognitive challenges in accepting feedback.
Computer Science,Education
What problem does this paper attempt to address?