All published articles of this journal are available on ScienceDirect.

RESEARCH ARTICLE

The Impact of AI-integrated Sport Blended Learning on Primary School Students' Sports Skills and Attitudes

The Open Sports Sciences Journal 23 July 2025 RESEARCH ARTICLE DOI: 10.2174/011875399X397619250721071324

Abstract

Introduction

This study examines the impact of AI-integrated Sport Blended Learning (AI-SBL) versus Traditional Face-To-Face Sport Learning (TFSL) on primary students' sport skills and attitudes. AI-integrated SBL provides personalized learning proven to enhance sports performance and PE engagement. We specifically investigate how AI-SBL enhances skills and promotes positive attitudes towards sports learning.

Methods

This study selected 94 students aged 8-9 years from Xinghu Primary School in Nantong City, Jiangsu Province. The subjects were divided into an experimental group (48 students) and a control group (46 students) using a computer-generated random allocation method. After confirming that there was no individual difference between the two groups through baseline data analysis, the experimental group implemented AI-integrated sport-blended learning teaching. In contrast, the control group conducted a comparative study using the traditional teaching method. Two Artificial Intelligence (AI) tools were used: Exercise Load Detection System (ELMS), which monitors exercise compliance via AI-driven tracking and feedback in PE classes. Daily Jump Rope App (DJR): A mobile app connecting in-class/after-class learning through real-time feedback, video tutorials, and fitness compensation. Sport skills (Vital capacity, 50-meter run, Sit-up-and-bend, 1-minute Jump rope) were analyzed via pre-post SBL data. Attitudes were assessed through four dimensions: sports attitude, learning purpose, learning time, and learning evaluation.

Results

AI-SBL significantly outperformed TFSL in skill development (all p<0.01; mean differences: Vital capacity = -1545, 50-meter run = 9.26, Sit-up-and-bend =13.37, 1-minute Jump rope = 141) and attitude improvement. Attitude analysis revealed consistently high evaluations: learning time (M = 4.27), sports attitude (M = 4.19), learning evaluation (M = 4.18), and learning purpose (M = 4.11), with SBL receiving more positive feedback.

Discussion

In physical education teaching, the AI-supported SBL model is significantly superior to TFSL in enhancing students' motor skills and learning attitudes. SBL integrates pre-class micro-videos, in-class ELMS, and DJR to achieve interaction between in-class and out-of-class activities. Its personalized and data-driven characteristics contrast sharply with traditional physical education teaching (TFSL). Empirical evidence shows that this model effectively improves the Vital capacity, 50-meter run, Sit-up-and-bend, and 1-minute Jump rope performance of primary school students. These skill advancements have further promoted a positive transformation in students' attitudes towards physical education, as evidenced by increased time investment and improved cognitive levels.

Conclusion

AI-SBL achieved superior sport performance and attitude outcomes compared to TFSL. This demonstrates AI-SBL's transformative potential in sports education methodology, providing valuable references for future research.

Keywords: Artificial intelligence, Sport blended learning, Sport skills, Sport attitudes, Exercise load detection system, Daily jump rope App.
Fulltext HTML PDF
1800
1801
1802
1803
1804