Ask AI: give 4 reasons why normality tests may not be as useful (such as the Shapiro-Wilks test)

1. Normality tests assume that the underlying data follows a normal distribution; however, in some cases the data may follow a different distribution that is not normal. 2. Normality tests assume that the data points are independent, however, in some cases dependent data points may exist which could result in misleading results. 3. Normality tests assume that the data points are randomly sampled, however, in some cases non-random sampling could occur which could also lead to misleading results. 4. Normality tests are generally considered to be “goodness of fit” tests and do not really provide any indication of a true underlying distribution which could lead to confusion.