Department of Biostatistics and Bioinformatics, Rollins School of Public Health, Emory University, Atlanta, GA 30322, USA.
Department of Applied Health Science, School of Public Health-Bloomington, Indiana University, Indianapolis, IN 46805, USA.
Int J Environ Res Public Health. 2023 Dec 1;20(23):7131. doi: 10.3390/ijerph20237131.
We conducted a critical review of the article "Effects on Children's Physical and Mental Well-Being of a Physical-Activity-Based School Intervention Program: A Randomized Study", published in the in 2023 as part of the Special Issue "Psychomotricity and Physical Education in School Health". We identified multiple mistakes in the statistical analyses applied. First, the authors claim to have found a statistically significant association between the proposed intervention and change in body composition (body mass index (BMI) percentiles, relative fat mass, and BMI classes) by way of exhibiting differences in nominal significance between the pre- and post-intervention changes the control and intervention groups, instead of exhibiting a significant difference groups. Furthermore, the analysis described fails to account for clustering and nesting in the data. The reporting of the statistical methods and results include multiple elements that are variously incorrect, incoherent, or impossible. Revised statistical analyses are proposed which can render the study's methods valid and its results substantiated, whereas the current methods and results are invalid and unsubstantiated, respectively.
我们对发表在 2023 年《特殊问题:心理运动学和学校健康体育教育》中的文章“基于身体活动的学校干预计划对儿童身心健康的影响:一项随机研究”进行了批判性评价。我们发现应用的统计分析中存在多个错误。首先,作者声称通过展示对照组和干预组之间干预前后变化的名义显著性差异,发现了所提出的干预措施与身体成分(体重指数(BMI)百分位数、相对脂肪量和 BMI 类别)变化之间存在统计学关联,而不是显示组间存在显著差异。此外,所描述的分析未能考虑数据中的聚类和嵌套。统计方法和结果的报告包含多个元素,这些元素在不同程度上是不正确的、不连贯的或不可能的。我们提出了修订后的统计分析方法,这些方法可以使研究的方法有效,并使结果得到证实,而当前的方法和结果是无效的,也是没有依据的。