The results of the Mann-Whitney test will not always match the results reported by older (5 and earlier) versions of Prism for two reasons.

•Exact vs. approximate P values. When samples are small, Prism computes an exact P value. When samples are larger, Prism computes an approximate P value. This is reported in the results. Prism is much (much!) faster at computing exact P values, so will do so with much larger samples. It does the exact test whenever the smaller group has fewer than 100 values.

•How to handle ties? If two values are identical, they tie for the same rank. Prism 6 and later, unlike most programs,  computes an exact P value even in the presence of ties. Prism 5 an earlier versions always computed an approximate P value, and different approximations were used in different versions. Details. Bergmann and colleagues (1) published a challenging example data set (with two variations) for the Mann-Whitney test, and Prism matches the results of StatXact, which they considered to be the gold standard (correct).

•Bugs. This page lists fixed bugs in Prism regarding this test.