医学科学研究中常用的t检验、卡方检验等检验方法,都是为了解决有无显著性差异这类问题的。鉴于t检验、卡方检验等概念比较抽象,计算方法也比较复杂,并且这些方法都是在母体遵从正态分布或近似正态分布的前提下才能使用,致使在应用时每感不便。本文介绍一种简易实用的显著性检验方法,它不受正态母体条件的限制,这就是秩和检验。因此,当在母体情况不清楚的小样本情况下,秩和检验就显示出它的优越性。其次,秩和检验的计算方法特别简单,所以即使在正态母体情况时,有时也宁可牺牲一点精度,而采用此法。什么叫做秩和检验呢?秩就是指从两个母体A(x)和B(x)中,分别随机抽取n_1和n_2个随村变量a_1,a_2,a_3,……a_(n_1)和b_1,b_2,b_3,……b_(n_2),再将它们按大小次序混合排列,这个重新排列的次序就叫做秩。我们假设母体A(x)=B(x),如果这个假设成立, Commonly used in medical science research t test, chi-square test and other testing methods are to solve the problem of whether there is any significant difference. In view of t-test, chi-square test and other concepts more abstract, the calculation method is also more complicated, and these methods are in the mother to follow the normal distribution or approximate normal distribution of the premise to use, resulting in each inconvenience in the application. This article presents a simple and practical significance test method, which is not subject to the restrictions of normal maternal conditions, which is the rank sum test. Therefore, the rank sum test shows its superiority in the case of a small sample whose maternal condition is not clear. Secondly, the calculation of rank sum test is particularly simple, so even in the case of normal maternal, and sometimes would rather sacrifice a little accuracy, and use this method. What is the rank sum test? Rank means that n_1 and n_2 variables a_1, a_2, a_3, ... a_ (n_1) and b_1 are extracted randomly from two parents A (x) and B b_2, b_3, ... b_ (n_2), and then mix them in order of size, the rearranged order is called rank. We assume that the parent A (x) = B (x), if this assumption holds,