A vice president for marketing recruits 20 college graduates for management training. Each of the 20 individuals is randomly assigned, 10 each, to one of the two groups. A "traditional" method of training (T) is used in one group, and an "experimental" method (E) is used in the other. After the graduates spend six months on the job, the vice president ranks them on the basis of their performance, from 1 (worst) to 20 (best), with the following results (stored in the file TestRank);

T: 1 2 3 5 9 10 12 13 14 15

E: 4 6 7 8 11 16 17 18 19 20

Is there evidence of a difference in the median performance between the two methods? (Use x = 0. 05).