This preview has intentionally blurred sections. Sign up to view the full version.
View Full Document
Unformatted text preview: 2, 2, 4, 3, 4, 5 H : m N = m A H : m N > m A three 1’s five 2’s seven 3’s eight 4’s seven 5’s one N, two A one N, four A four N, three A three N, five A six N, one A would have been ranked 1, 2, 3 would have been ranked 4, 5, 6, 7, 8 would have been ranked 9, 10, 11, 12, 13, 14, 15 would have been ranked 16, 17, 18, 19, 20, 21, 22, 23 would have been ranked 24, 25, 26, 27, 28, 29, 30 average = 2 average = 6 average = 12 average = 19.5 average = 27 W N = 1 × 2 + 1 × 6 + 4 × 12 + 3 × 19.5 + 6 × 27 = 276.5. W A = 2 × 2 + 4 × 6 + 3 × 12 + 5 × 19.5 + 1 × 27 = 188.5. Note that 276.5 + 188.5 = 465 = ( ) 2 1 30 30 + × . Z = 12 31 15 15 2 31 15 5 . 276 ⋅ ⋅ ⋅= 1.825 . pvalue = P ( Z ≥ 1.825 ) = 0.034 . pvalue < α Reject H . OR Z = 12 31 15 15 2 31 15 5 . 188 ⋅ ⋅ ⋅= – 1.825 . pvalue = P ( Z ≤ – 1.825 ) = 0.034 . pvalue < α Reject H ....
View
Full Document
 Fall '11
 STEPHANOV
 Harshad number, Self number, Statistical hypothesis testing, Statistical significance, new painkiller

Click to edit the document details