PCBs have been in use since 1929, mainly in the electrical industry, but it was not until the 1960s that they were found to be a major environmental contaminant. In the paper "The ratio of DDE to PCB concentrations in Great Lakes herring gull eggs and its use in interpreting contaminants data,"
† researchers report on the following study. Thirteen study sites from the five Great Lakes were selected. At each site, 9 to 13 herring gull eggs were collected randomly each year for several years. Following collection, the PCB content was determined. The mean PCB content at each site is reported in the following table for the years 1982 and 1996.
|
Site |
Year |
1 |
2 |
3 |
4 |
5 |
6 |
7 |
8 |
9 |
10 |
11 |
12 |
13 |
1982 |
61.48 |
64.47 |
45.50 |
59.70 |
58.81 |
75.86 |
71.57 |
38.06 |
30.51 |
39.70 |
29.78 |
66.89 |
63.93 |
1996 |
13.99 |
18.26 |
11.28 |
10.02 |
21.00 |
17.36 |
28.20 |
7.30 |
12.80 |
9.41 |
12.63 |
16.83 |
22.74 |
A potential criticism of analyzing these data as if they were two independent samples is that the measurements taken in 1996 were taken at the same site as the measurements taken in 1982. Thus, there is the possibility that there will be a strong positive correlation between the pair of observations at each site.
(a) Plot the pairs of observations in a scatterplot with the 1982 values on the horizontal axis and the 1996 values on the vertical axis.
0,0
5
10
15
20
25
30
35
40
45
50
55
60
65
70
75
80
2
4
6
8
10
12
14
16
18
20
22
24
26
28
30
Graph LayersToggle Open/Closed
- After you add an object to the graph you can use Graph Layers to view and edit its properties.
Does there appear to be a positive correlation between the pairs of measurements?
(b) Compute the correlation coefficient between the pair of observations.
r =
Does this value confirm your observations from the scatterplot?
(c) Answer the following questions using a paired data analysis.
Are your conclusions different from the conclusions you reached treating the data as two independent samples?