Which plot type in PROC UNIVARIATE allows for comparison against theoretical distributions?

The quantile-quantile (Q-Q) plot is specifically designed for comparing the quantiles of the sample data against the quantiles of a theoretical distribution, such as the normal distribution. This graphical tool is invaluable for assessing how closely the distribution of the observed data aligns with the expected distribution. When the data perfectly follows the theoretical distribution, the points will lie on a straight diagonal line. Deviations from this line indicate how the actual data diverges from the theoretical model.

In the context of PROC UNIVARIATE, this plot allows researchers to visually assess the goodness-of-fit of their data to a specified theoretical distribution. By examining this plot, statisticians can identify whether their data exhibits characteristics (like skewness or kurtosis) that may suggest it does not conform to the assumed distribution, guiding further analytical decisions.

Other plot types mentioned serve different purposes: scatter plots are typically used to visualize relationships between two variables, box plots summarize data distributions with their quartiles, and bar charts represent categorical data. Therefore, among the options available, the Q-Q plot is uniquely suited for comparing with theoretical distributions, making it the correct choice.