论文部分内容阅读
为了规范体育科学研究方法,提高多元统计分析的科学性,针对统计描述、统计推断和综合评价3个体育科学研究领域中存在的典型问题,分析了问题的性质和原因,提出了解决问题的思路和具体建议。指出:在统计描述工作中,要明确描述对象,确定描述的内容,合理选择和使用统计指标,重视描述性评价;对于多元均值的推断,需要明确统计总体和个体,理清需要推断的问题,运用多元检验统计量检验多元均值向量,运用联合置信区间检验各个分量及其线性组合;关于体育综合评价,建议运用主成分或因子分析法分析评价指标集的内部结构,采用专家法对主成分或因子赋权。
In order to standardize the scientific research methods of sports and improve the scientificity of multivariate statistical analysis, this paper analyzes the typical problems existing in the field of sports science research in terms of statistical description, statistical inference and comprehensive evaluation, analyzes the nature and causes of the problems, and puts forward some solutions to the problems And specific suggestions. It points out that in the process of statistical description, it is necessary to clearly describe the object, determine the content of the description, reasonably select and use the statistical indicators, and emphasize on the descriptive evaluation. For the estimation of the multivariate means, the overall population and the individual need to be clearly defined, The multivariate test statistic is used to test the multivariate mean vector, and the joint confidence interval is used to test each component and its linear combination. For the comprehensive evaluation of sports, it is suggested to analyze the internal structure of the evaluation index set by principal component or factor analysis, Factor weighting.