EXAMPLE 6.31

Constructing a power curve. Example 6.30 considered one specific alternative, μ = 15.50. Often, it is helpful to consider the power for a range of alternatives. Fortunately, most statistical software saves us from having to do these calculations manually. Figure 6.18 shows Minitab output for the power over the range 15.00 ppm to 15.80 ppm. The power calculation of Example 6.30 is represented by a dot on the curve at a difference of 15.50 − 15.00 = 0.50. This curve is very informative. We see that with a sample size of three, the power is greater than 80% only for differences larger than about 0.48. If it is important to detect differences less than this, the Deely Laboratory needs to consider ways to increase the power.