A cattle feedlot operator has collected data for 40 matched pairs of cattle showing weight gain on two different feed supplements. His purpose in collecting the data is to determine whether there is a difference in the median weight gains for the two supplements. He has no preconceived idea about which supplement might produce higher weight gain. He wishes to test using an a = 0.05 level.

Assuming that the T-value for these data is 480, what should the operator conclude concerning which supplement produces higher weight gain? Use the large-sample Wilcoxon matched-pairs signed rank test normal approximation. Conduct the test using a p-value approach.