如何正确运用Z检验——总体方差已知时两算术均值比较一般差异性Z检验及SAS实现  

How to use Z test correctly——comparison of two arithmetic mean values for the general difference Z test under the condition of known the population variance and the SAS implementation

作　　者：胡良平 Hu Liangping(Graduate School,Academy of Military Sciences PLA China,Beijing 100850,China;Specialty Committee of Clinical Scientific Research Statistics of World Federation of Chinese Medicine Societies,Beijing 100029,China)

机构地区：[1]军事科学院研究生院,北京100850 [2]世界中医药学会联合会临床科研统计学专业委员会,北京100029

出　　处：《四川精神卫生》2020年第5期422-426,共5页Sichuan Mental Health

摘　　要：本文的目的是介绍总体方差已知时两算术均值比较一般差异性Z检验及SAS实现。首先,介绍了选用Z检验的前提条件;其次,介绍了单组设计和成组设计条件下进行均值比较的Z检验统计量,并介绍了上单侧检验、下单侧检验和双侧检验时所对应的拒绝域;最后,结合两个实例,基于SAS软件实现3种Z检验,对输出结果做出了解释。并针对如何合理选择单侧检验还是双侧检验给予说明。The purpose of this paper was to introduce the comparison of two arithmetic mean values for the general difference Z test under the condition of known the population variance and the SAS implementation.Firstly,the preconditions of the Z test was introduced.Secondly,the Z statistics were given for the comparison of two arithmetic mean values for the general difference Z test under the designs of the single group and two groups.Some basic concepts,such as upper one-sided test,lower one-sided test,twosided test and the rejection range,were introduced.Lastly,three kinds of Z tests were performed based on the SAS software combined with two real examples,and the output results were also explained.Furthermore,the necessary explanation was given about how to select one-sided test or two-sided test rationally.

关 键 词：正态分布 Z检验 单侧检验 双侧检验 拒绝域 

分 类 号：R195.1[医药卫生—卫生统计学]