Which graph is best used to show the distribution of numerical data?

A histogram is the most suitable graph for showing the distribution of numerical data because it provides a visual representation of data frequency across different ranges or bins. Each bar in a histogram represents the frequency of data points that fall within a specific interval of numerical values. This allows viewers to easily observe patterns such as skewness, modality (the number of peaks), and the overall shape of the distribution.

For instance, if you want to understand how students' test scores are distributed over a range from 0 to 100, a histogram will show how many students scored within various score intervals, allowing you to see whether most students performed well, poorly, or if there is a normal distribution.

In contrast, other graph types serve different purposes; a line graph is typically used to demonstrate trends over time, connecting individual data points with lines. A box plot summarizes data by indicating the median, quartiles, and potential outliers, providing a compact summary rather than a detailed distribution view. A pie chart displays proportions of categories within a whole and is not suitable for presenting numerical distributions at all. Therefore, the histogram stands out as the optimal choice for illustrating the distribution of numerical data.