School of Physical Education and Sport, Shandong Normal University, Jinan, China.
School of Physical Education and Sport, Shandong Agricultural University, Tai'an, China.
Child Care Health Dev. 2024 Jul;50(4):e13275. doi: 10.1111/cch.13275.
This study aimed to investigate the psychometric characteristics of the Pictorial Scale of Perceived Movement Skill Competence (PMSC) in Chinese children, utilizing the principles of item response theory (IRT).
A total of 284 children (147 boys), aged 4-10 years (M = 7.39, SD = 2.18), from four schools were recruited for testing, adhering to the principle of random sampling. The Rasch model was applied to evaluate the rating scale validity, local independence, scale dimensionality, person and item fit statistics, and differential item functioning (DIF).
(1) The rating scale validity was confirmed by the monotonically increasing trend in both average item difficulty measures and Andrich thresholds. (2) However, the observation that standardized residual correlations for all items remained below 0.3 unequivocally confirmed the presence of local independence among the variables. (3) The scale's dimensions were substantiated, with Rasch measures explaining 40.6% of total variance, exceeding the 40% threshold. Additionally, the first contrast had an eigenvalue of 1.8 and explained 8.80% of the variance, both exceeding the minimum 3:1 ratio. Consequently, the unidimensionality assumption of the item was supported. (4) Person and item fit statistics indicated requisite reliability and separation, with all achieving suitable infit and outfit measures, and point-biserial correlations exceeding 0.55. (5) Regarding gender DIF, it was observed that the majority of items did not exhibit statistically significant gender DIF effects (p > 0.05) except item 8 (Bounce a ball, t = -2.74, p < 0.01, DIF = -0.51).
Rasch analysis revealed that the instrument exhibited strong scale validity among Chinese children, adhering to the Rasch model's unidimensionality and local independence principles, with robust person and item reliability and separation. All items demonstrated a good fit. Additionally, the significant gender difference in item 8 (Bounce a ball) warrants further consideration. In conclusion, PMSC is suitable for assessing the PMSC of Chinese children.
本研究旨在运用项目反应理论（IRT）的原理，探讨中文版图画运动技能感知能力量表（PMSC）在中国儿童中的心理计量学特征。
本研究共招募了来自四所学校的 284 名 4-10 岁（M=7.39，SD=2.18）的儿童进行测试，遵循随机抽样原则。应用 Rasch 模型评估评分量表的效度、局部独立性、量表维度、个体和项目拟合统计以及项目间差异功能（DIF）。
（1）平均项目难度测量和 Andrich 阈值的单调递增趋势证实了评分量表的效度。（2）然而，所有项目的标准化残差相关性均低于 0.3，这一事实明确证实了变量之间存在局部独立性。（3）量表维度得到了证实，Rasch 测量解释了总方差的 40.6%，超过了 40%的阈值。此外，第一个对比的特征值为 1.8，解释了 8.80%的方差，均超过了最小 3:1 的比例。因此，项目的单维性假设得到了支持。（4）个体和项目拟合统计数据表明需要可靠的分离，所有项目的 infit 和 outfit 测量均达到合适的标准，点双列相关系数超过 0.55。（5）关于性别 DIF，观察到大多数项目没有表现出统计学上显著的性别 DIF 效应（p>0.05），除了项目 8（Bounce a ball，t=-2.74，p<0.01，DIF=-0.51）。
Rasch 分析表明，该工具在中国儿童中表现出较强的量表效度，符合 Rasch 模型的单维性和局部独立性原则，具有可靠的个体和项目可靠性和分离性。所有项目都表现出良好的拟合度。此外，项目 8（Bounce a ball）在性别上的显著差异需要进一步考虑。总之，PMSC 适合评估中国儿童的 PMSC。
