摘要
采用Python语言,分别使用主成分分析法和因子分析法,对我国30个省市中医药服务能力进行了排名和聚类分析,从而为科学客观评价我国中医药服务能力提供方法参考。结果表明,两种方法对应的30省市能力排名结果非常相似:排名结果完全的一致省市有8个,排名仅相差1位的有13个省市,相差2位的有7个省市。但聚类分析结果有些差别:四类中仅有一类完全一致,其他三类均有差别,个别类别差异较大。总体来看,两种方法均适用能力评价实证分析。因子分析假设公因子互不相关,而主成分肯定互不相关,且选取的主成分和公因子各自代表的原始变量信息比例不同,这些可能是导致二者分析结果存在差异的主要原因。本文未从原始变量相关性来分析两种方法的异同,这是未来研究努力的方向。
Using Python language,the service ability of traditional Chinese medicine in 30 provinces in China was ranked and cluster analyzed based on Principal Component Analysis Method and Factor Analysis Method in order to provide method reference for scientific and objective evaluation of Chinese medicine service ability in China.The results show that the capacity ranking results of 30 provinces corresponding to the two methods are very similar:the ranking results of 8 provinces are completely consistent,,13 provinces with only 1 place difference,and 7 provinces with 2 places difference.But there are some differences in the results of cluster analysis:only one of the four categories is completely consistent,the other three are different,and individual categories are quite different.Generally speaking,both methods are applicable to the empirical analysis of capacity evaluation.Factor analysis assumes that common factors are uncorrelated with each other,while principal components are definitely not correlated with each other,and the proportion of original variables represented by the selected principal components and common factors are different,which may be the main reasons for the differences between the two analysis results.This paper does not analyze the similarities and differences between the two methods from the original variable correlation,which is the direction of future research.
作者
赵敏
李军山
ZHAO Min;LI Jun-shan(Nanchang Hangkong University,Nanchang 330063,China;Jiangxi University of Traditional Chinese Medicine,Nanchang 330004,China)
出处
《江西中医药大学学报》
2020年第6期103-106,共4页
Journal of Jiangxi University of Chinese Medicine
基金
江西省教育厅教学改革项目(JXJG-19-12-16)
江西中医药大学教学改革项目(2017jzyb-5)。
