Suppose that you want to test the hypothesis that mothers with low socioeconomic status (SES) deliver babies whose birthweights are lower than "normal" at the 5% significance level.To test this hypothesis, you want to conduct a study in which you collect data from full-term live-born deliveries from various maternity wards of hospitals in low SES areas. It is assumed that the standard deviation of birthweights is 24 oz. A national survey has shown that "normal" birthweight is 120 oz.

You want to consider being able to detect differences of 5 oz, 10 oz, or 15 oz below normal birthweight. You decide to construct a power curve for each one of these differences where you look at power as a function of sample size. These three power curves are provided below.