The fraction p of common words between texts given in Table 1.

Now p is defined as follows. Let n(i) and n(ij) be, respectively, the number of different words in text i and the number of common words in texts i and j. We define: p(ij) = n(ij)/(n(i) + n(j) − n(ij)), where 0 ≤ p(ij) ≤ 1. This is the number of common words divided over the number of all different words in texts i and j. As seen from the data below, analogues of Eqs (21)–(23) hold with 1 − p(ij) instead of ρλ(ij).