Use a Wilcoxon Signed Rank Test of the alternate hypothesis H1 : µ does not equal 7.39 and the following data:
7.02 7.35 7.34 7.17 7.28 7.77 7.09
7.22 7.45 6.95 7.40 7.10 7.32 7.14
to report the observed level of statistical significance.
vs
Define
Di =Xi -3.79
Since the addition is community WLG we assume that |D1|< |D2|<...<|Dn|
Rank |Di| =i
Define Z(i)=1 if the difference whose absolute value has rank i is positive
0 o.w
we can write
Then we arrange the value in ascending order
6.95, 7.02, 7.09, 7.10, 7.14, 7.17, 7.22, 7.28, 7.32, 7.34,7.35,7.40, 7.45, 7.77 and Di =0.44, 0.37, 0.3,0.29, 0.25, 0.22, 0.17, 0.11, 0.07, 0.05, 0.04, 0.01, 0.06, 0.38 respectively
i | Di | sign |
1 | 0.01 | + |
2 | 0.04 | - |
3 | 0.05 | - |
4 | 0.06 | + |
5 | 0.07 | - |
6 | 0.11 | - |
7 | 0.17 | - |
8 | 0.22 | - |
9 | 0.25 | - |
10 | 0.29 | - |
11 | 0.3 | - |
12 | 0.37 | - |
13 | 0.38 | + |
14 | 0.44 | - |
W+= (1+4+13)
=18
E(W+)= n(n+1)/4
=14(14+1)/4
=42
Var(W+) =n(n+1)(2n+1)/24
=14(14+1)(2*14+1)/24
=253.75
test statistic
=(18-42)/15.93
=-1.51
p value 0.13 > 0.05 so we accept null hypothesis
Get Answers For Free
Most questions answered within 1 hours.