Which statistical analysis would not be used if the data is not normally distributed?

When data is not normally distributed, parametric statistical analyses are typically not appropriate because these analyses assume that the data follows a normal distribution. Parametric tests, such as t-tests and ANOVA, rely on this assumption to perform analyses and draw conclusions. When the normality assumption is violated, these tests can produce misleading results, making it crucial to select appropriate methods suited for the data's distribution characteristics.

In contrast, nonparametric tests do not assume a normal distribution and are more flexible in handling various types of data. They can provide reliable results even when the data does not meet normality assumptions. Descriptive statistics can always be calculated, as they summarize and describe the characteristics of the dataset without making assumptions about its distribution. Inferential statistics may also include nonparametric procedures, which can still draw conclusions from non-normally distributed data. Thus, parametric analysis is the correct choice when identifying the type of analysis that should not be used with non-normally distributed data.