论文部分内容阅读
多个样本均数之间差异比较,经过方差分析得出在统计学上有显著意义的结论(P<0.05),只是说明各组之间总的来看差异有显著意义,但并不意味着任何二组之间差别均有显著意义.如想进一步推断哪二组之间差别有显著意义,哪二组之间差别无显著意义,则需进行两两比较.此时又有二种情况,一种是每二个样本均数之间都要比较过来,可选用Newman-keuls q检验;另一种情况是比较的各组中有一个设定的对照组,其它各组都和它进行比较.可选用Dunnett检验.下面分别用两个实例说明这两种检验方法的计算步骤.
A statistically significant difference (P <0.05) was obtained by means of ANOVA between the multiple sample mean differences, only indicating a significant difference between the groups in general but this does not imply Any difference between any two groups have significant significance.If you want to further infer which two groups have significant differences, the difference between the two groups is not significant, you need to compare every two.At this time there are two cases, One is to compare the average of every two samples, Newman-Keuls q test can be used; the other case is to compare a set of control groups in each group, the other groups are compared with it Can choose Dunnett test.The following two examples are used to illustrate the calculation of these two test methods.