Statistics for Environmental Engineers

Скачать в pdf «Statistics for Environmental Engineers»

Case Study: Interlaboratory Study of Dissolved Oxygen


An important procedure in certifying the quality of work done in laboratories is the analysis of standard specimens that contain known amounts of a substance. These specimens are usually introduced into the laboratory routine in a way that keeps the analysts blind to the identity of the sample. Often the analyst is blind to the fact that quality assurance samples are included in the assigned work. In this example, the analysts were asked to measure the dissolved oxygen (DO) concentration of the same specimen using two different methods.


Fourteen laboratories were sent a test solution that was prepared to have a low dissolved oxygen concentration (1.2 mg/L). Each laboratory made the measurements using the Winkler method (a titration) and the electrode method. The question is whether the two methods predict different DO concentrations. Table 17.1 shows the data (Wilcock et al., 1981). The observations for each method may be assumed random and independent as a result of the way the test was designed. The differences plotted in Figure 17.1 suggest that the Winkler method may give DO measurements that are slightly lower than the electrode method.





TABLE 17.1


Dissolved Oxygen Data from the Interlaboratory Study


Laboratory


1


2


3


4


5


6


7


8


9


10


11


12


13

Скачать в pdf «Statistics for Environmental Engineers»