Background: The importance of physical education in higher education is widely recognized, as it improves not only exercise attitudes and motor skills but also physical fitness, social skills, and academic performance. However, physical education courses in Chinese colleges and universities face various constraints, such as a shortage of teachers, limited teaching methods, and insufficient resources, resulting in low student motor skills, negative attitudes toward sports, and low participation rates. This study explores the effectiveness of a blended learning model, which integrates traditional face-to-face instruction with online learning components, in improving university students' exercise attitudes and basketball skills.

Methods: The research was conducted in 2022 at Luoyang Normal University in China, utilizing a cluster randomized controlled trial (CRCT) with 78 healthy first-year university students. Participants were randomly assigned to either the experimental group (blended learning) or the control group (traditional learning), and the intervention lasted 16 weeks. Exercise attitudes were assessed via the Exercise Attitude Scale, whereas basketball skills were evaluated via set shot and half-court dribbling and shooting tests.

Results: Both instructional models improved students' exercise attitudes and basketball skills; however, the blended learning model demonstrated significantly superior outcomes. Effect sizes (d) ranging from 0.57 to 1.92 indicated that the experimental group showed greater improvements in behavior attitude, target attitude, behavior cognition, behavior intention, emotional experience, behavior control, and subjective standards. In basketball skills, the experimental group outperformed the control group in set shots (d = 0.56) and half-court dribbling and shooting (d = 0.46).

Conclusion: Compared with traditional methods, blended learning significantly enhances university students' exercise attitudes and basketball skills. Future research should explore the long-term effects and underlying mechanisms of blended learning in physical education, involving larger and more diverse samples to validate these findings.