Abstract:

【Objective】 The former paper had proposed the rank-sum testing method for multi-trait comprehensive evaluation, which not only can make the multi-trait comprehensive ranking, but also test the significance of difference between evaluation objects and the average level. In this paper, a rank-sum-difference testing method for multi-trait comprehensive evaluation was studied to implement the multiple comparisons among evaluation objects. 【Method】 The theoretical distribution of rank-sum-difference, based on the theoretical distribution of rank-sum, was firstly deduced by using combinatorial mathematics.【Result】 Furthermore, it was used to obtain the critical values for the rank-sum-difference testing in multi-trait comprehensive evaluation. 【Conclusion】 The proposed rank-sum-difference testing method is so simple and convenient that it can be easily used to realize multiple comparisons among various evaluation objects.

Key words: comprehensive evaluation, multiple comparisons, rank-sum-difference testing