The following table presents samples of hourly yields for catalysts XA-100 and ZB-200. We analyzed these data using a two independent sample t test in Example 10.4 (page 405).
Catalyst XA-100Catalyst ZB-200
801...........................752
814...........................718
784...........................776
836...........................742
820...........................763

a. Use a Wilcoxon rank sum test and the Excel add-in (MegaStat) output to test for systematic-differences in the yields of the two catalysts. Use α = .05.
b. The p-value on the MegaStat output has been calculated by finding twice the area under the standard normal curve to the right of

Here we have used a continuity correction and changed T1 = 40 to 39.5. Verify the calculations of µT, σT, and the p-value.
c. Assume that the second yield for catalyst ZB-200 in the above table is invalid. Use the remaining data to determine if we can conclude that the XA-100 yields are systematically higher than the ZB-200 yields. Set α = .05.

Box Plots of Yield by Catalyst 820- 770 720- XA-100 ZB-200 Catalyst Wilcoxon-Mann/Whitney Test n sum of ranks 10a XA-100 27.50 expected value ZB-200 4.79 standard deviation 0122 p-value (two-tailed) total 2.51 z 10 39.5-27.5 4.79 = 2.51 39.5-H7