PDF(Pubmed)
Abstract：
The research analyzed the effect of weekly training plans, physical training frequency, AI-powered coaching systems, virtual reality (VR) training environments, wearable sensors on developing technical tennis skills, with and personalized learning as a mediator. It adopted a quantitative survey method, using primary data from 374 young tennis players. The model fitness was evaluated using confirmatory factor analysis (CFA), while the hypotheses were evaluated using structural equation modeling (SEM). The model fitness was confirmed through CFA, demonstrating high fit indices: CFI = 0.924, TLI = 0.913, IFI = 0.924, RMSEA = 0.057, and SRMR = 0.041, indicating a robust model fit. Hypotheses testing revealed that physical training frequency (β = 0.198, p = 0.000), AI-powered coaching systems (β = 0.349, p = 0.000), virtual reality training environments (β = 0.476, p = 0.000), and wearable sensors (β = 0.171, p = 0.000) significantly influenced technical skills acquisition. In contrast, the weekly training plan (β = 0.024, p = 0.834) and personalized learning (β = -0.045, p = 0.81) did not have a significant effect. Mediation analysis revealed that personalized learning was not a significant mediator between training methods/technologies and acquiring technical abilities. The results revealed that physical training frequency, AI-powered coaching systems, virtual reality training environments, and wearable sensors significantly influenced technical skills acquisition. However, personalized learning did not have a significant mediation effect. The study recommended that young tennis players\' organizations and stakeholders consider investing in emerging technologies and training methods. Effective training should be given to coaches on effectively integrating emerging technologies into coaching regimens and practices.
摘要：
研究分析了每周培训计划的效果,体育锻炼频率,人工智能驱动的教练系统，虚拟现实（VR）训练环境，可穿戴传感器发展技术网球技能，以个性化学习为中介。它采用了定量调查的方法，使用374名年轻网球运动员的主要数据。使用验证性因子分析(CFA)评估模型适合度，同时使用结构方程模型(SEM)对假设进行了评估。通过CFA确认了模型的适用性，表现出较高的拟合指数：CFI=0.924，TLI=0.913，FI=0.924，RMSEA=0.057，SRMR=0.041，表明模型拟合稳健。假设检验表明，体育锻炼频率(β=0.198，p=0.000)，AI驱动的教练系统（β=0.349，p=0.000），虚拟现实训练环境(β=0.476，p=0.000)，可穿戴传感器(β=0.171，p=0.000)显著影响技术技能获取。相比之下,周训练计划(β=0.024,p=0.834)和个性化学习(β=-0.045,p=0.81)效果不显著。中介分析显示，个性化学习并不是训练方法/技术与获得技术能力之间的重要中介。结果显示，体育锻炼的频率，人工智能驱动的教练系统，虚拟现实训练环境，和可穿戴传感器显著影响技术技能的获取。然而,个性化学习没有显著的中介效应。该研究建议年轻网球运动员的组织和利益相关者考虑投资于新兴技术和训练方法。应该对教练进行有效的培训，以将新兴技术有效地整合到教练方案和实践中。
