水迷宫重复测量数据的方差分析及其在SPSS中的实现  

Analysis of variance of repeated data measured by water maze with SPSS

作　　者：邱宏[1] 金国琴[2] 金如锋[1] 赵伟康[3] 

机构地区：[1]上海中医药大学基础医学院预防医学与卫生统计学教研室,上海201203 [2]上海中医药大学基础医学院生物化学教研室,上海201203 [3]上海中医药大学老年医学研究所,上海201203

出　　处：《中西医结合学报》2007年第1期101-105,共5页Journal of Chinese Integrative Medicine

基　　金：国家自然科学基金重点资助项目(No.39830450)

摘　　要：目的:介绍用SPSS11.0实现对水迷宫重复测量资料的方差分析,以期对采用重复测量设计方案的临床和基础科研工作者提供可借鉴的统计学分析方法。方法:应用SPSS11.0中一般线性模型(generallinearmodel,GLM)的repeatedmeasures和multivariate过程对水迷宫重复测量数据进行重复测量方差分析和多元方差分析,并进行不同时间点和不同组间的两两比较。结果:首先通过球形检验(Mauchly’stestofsphericity)的结果判断重复测量数据之间是否存在相关性,如存在相关性(P≤0.05),宜进行多元方差分析,或采用Greenhouse-Geisser的校正结果;通过计算个体间(between-subject)变异,可分析处理因素有无效应;计算个体内(within-subject)变异,可分析时间因素有无效应,时间与处理因素之间有无交互效应;使用重复测量数据多重比较配对的t检验法(Bonferroni法),可进行每个分组每个时间点上作用的两两比较;使用多元方差分析的方法,可进行每个时间点上每个分组之间作用的两两比较。结论:对于重复测量数据的方差分析,宜采用一般线性模型的重复测量过程方法进行分析。SPSS统计软件易于实现重复测量数据的方差分析。Objective： To introduce the method of analyzing repeated data measured by water maze with SPSS 11. 0, and offer a reference statistical method to clinical and basic medicine researchers who take the design of repeated measures. Methods; Using repeated measures and multivariate analysis of variance （ANOVA） process of the general linear model in SPSS and giving comparison among different groups and different measure time pairwise. Results： Firstly, Mauchly＇s test of sphericity should be used to judge whether there were relations among the repeatedly measured data. If any （ P ≤0.05）, multivariate ANOVA should be taken next, or Greenhouse- Geisser corrected results should be taken. Treated effect could be evaluated by estimating between-subject variance. Repeated measurement effect or its interactive effect with treated group could be evaluated by estimating within-subject variance. The method of Bonferroni should be used to do pairwise comparisons of the repeatedly measured data in different measurement time of each treated group. With multivariate ANOVA, data in different treated group of each measurement time could be compared pairwise. Conclusion： The repeated measures process of the general linear model is suitable for variance analysis of repeatedly measured data. SPSS statistical package is available to fulfil this process.

关 键 词：方差分析 统计学 SPSS 线性模型 

分 类 号：R285.5[医药卫生—中药学]