In: Statistics and Probability
#33 Babita
Turbid water is muddy or cloudy water. Sunlight is necessary for most life forms; thus turbid water is considered a threat to wetland ecosystems. Passive filtration systems are commonly used to reduce turbidity in wetlands. Suspended solids are measured in mg/l. Is there a relation between input and output turbidity for a passive filtration system and, if so, is it statistically significant? At a wetlands environment in Illinois, the inlet and outlet turbidity of a passive filtration system have been measured. A random sample of measurements are shown below. (Reference: EPA Wetland Case Studies.)
Reading | 1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9 | 10 | 11 | 12 |
Inlet (mg/l) | 88.5 | 58.5 | 69.8 | 58.0 | 80.4 | 20.2 | 18.4 | 63.6 | 91.5 | 55.7 | 70.5 | 90.5 |
Outlet (mg/l) | 9.9 | 4.2 | 13.4 | 5.8 | 9.4 | 4.8 | 3.2 | 10.5 | 16.1 | 3.5 | 7.6 | 14.5 |
Use a 1% level of significance to test the claim that there is a monotone relationship (either way) between the ranks of the inlet readings and outlet readings.
(a) Rank-order the inlet readings using 1 as the largest data value. Also rank-order the outlet readings using 1 as the largest data value. Then construct a table of ranks to be used for a Spearman rank correlation test.
Reading | Inlet Rank x |
Oulet Rank y |
d = x - y | d2 |
1 2 3 4 5 6 7 8 9 10 11 12 |
Σd2 = |
(c) Compute the sample test statistic. (Use 3 decimal
places.)
a) From the given data
X | Y | Ranks of X | Ranks of Y | di = R(xi)-R(yi) | di^2 |
88.5 | 9.9 | 3 | 5 | -2 | 4 |
58.5 | 4.2 | 8 | 10 | -2 | 4 |
69.8 | 13.4 | 6 | 3 | 3 | 9 |
58 | 5.8 | 9 | 8 | 1 | 1 |
80.4 | 9.4 | 4 | 6 | -2 | 4 |
20.2 | 4.8 | 11 | 9 | 2 | 4 |
18.4 | 3.2 | 12 | 12 | 0 | 0 |
63.6 | 10.5 | 7 | 4 | 3 | 9 |
91.5 | 16.1 | 1 | 1 | 0 | 0 |
55.7 | 3.5 | 10 | 11 | -1 | 1 |
70.5 | 7.6 | 5 | 7 | -2 | 4 |
90.5 | 14.5 | 2 | 2 | 0 | 0 |
Total = | 40 |
c) Test for correlation Coefficient :
Test Statistic t = 5.332