箱线图

A boxplot is constructed from five values: the minimum value, first quartile, median (second quartile), third quartile, and maximum value. These values create a 5 number summary that visually represents the distribution and spread of your data, showing both central tendency and extreme values at a glance.

First, sort your data from low to high and find the 5-number summary. Draw a horizontal or vertical number line with a rectangular box extending from the first quartile to the third quartile. Place a line through the box at the median, then draw whiskers connecting the box ends to the minimum and maximum values.

In normal distributions, the median typically appears centered within the box. In skewed distributions, the median line shifts forward or backward within the box, indicating the direction and degree of skewness. This visual positioning quickly shows whether data is symmetrically distributed or leans toward higher or lower values.

The whiskers extend from the box ends to the smallest and largest data values, showing the full range of your dataset. They reveal how far extreme values are from the central 50 percent of data contained in the box, helping identify the overall spread and potential what are outliers in your distribution.

Boxplots allow side-by-side visual comparison of different datasets' distributions, medians, and spreads. For example, comparing goal-scoring performance across multiple World Cup seasons reveals how a team's performance changed over time, making patterns and differences immediately apparent without detailed numerical analysis.

Approximately the middle 50 percent of data falls inside the box, bounded by the first and third quartiles. This central box region represents the interquartile range, showing where half your data concentrates and providing a clear picture of data concentration around the median.

A boxplot gives a quick graphical image of data concentration by showing how tightly data clusters in the central box region. The whiskers simultaneously display how far extreme values extend from this concentration, allowing you to assess both typical data behavior and the presence of unusually distant observations.