论文部分内容阅读
多组等级分组资料比较,经秩和检验得出在统计学上有显著意义的结论(P≤0.05).只是说明比较各组间的差别总的来看有显著意义.但并不意味着其中任何两组间的差别均有显著意义.若想进一步推断究竟哪两组间差别有意义,哪两组间差别无意义.还是任何两组间差别均有意义.则要进行两两比较,所用方法国各组例数是否相等而有所不同.当各组例数都相等时可用Nemenyi提出的两两比较方法.当各组例数不等时可用Conover W.J提出的推广的t检验,下面分别通过实例,说明这两种检验方法.
The results of rank sum test showed significant statistical significance (P≤0.05), but only showed that the difference between the groups was significant, but it does not mean that Any difference between the two groups have significant significance.If you want to further infer whether the difference between the two groups meaningful, which is meaningless between the two groups, or any difference between the two groups are meaningful, then a pair of comparisons, the use of Method countries each group of cases are equal and different .When each group of cases are equal, Nemenyi proposed a pairwise comparison method .When the number of cases ranging from Conover WJ can be used to promote the t-test, the following respectively By example, the two test methods.