Question: Nonparametric tests, power, and variability. Serlin and Harwell (2004) analyzed the power of statistical designs when distributions are not normal. In their article, they reported, "It is well-known that for normally distributed errors parametric tests are optimal statistically, but perhaps less well-known is that when normality does not hold, nonparametric tests frequently possess greater statistical power than parametric tests" (Serlin & Harwell, 2004, p. 492). Explain why power can be greater using a nonparametric test when data are not normally distributed.