Box Plot
A box plot summarizes a numeric distribution with a median, quartiles, and whiskers. It is useful for comparing the center and spread of several groups without drawing every observation.
Read the box and the whiskers
The box runs from the first quartile (Q1) to the third quartile (Q3). It covers the middle half of the distribution. The line inside is the median. The interquartile range is Q3 minus Q1.
In this example, chat has a median response time of 12 minutes. Its middle half runs from 7 to 18 minutes. Email has a median of 30 minutes and a wider box.
The example uses minimum and maximum values for the whiskers. Another common convention extends them to the most extreme observations within 1.5 interquartile ranges of the box. Points outside those limits are shown separately. State which convention you use. See the NIST box plot guide.
What the summary leaves out
Different distributions can have similar boxes. A box plot does not show every cluster or peak, and the median alone does not establish a statistically significant difference.
Include group sizes. For small samples, show the individual observations as well. Use a histogram when the shape of a distribution is the main question.
Compare on the same scale
Use the same unit and axis range across groups. An observation beyond a whisker is a reason to investigate, not an automatic reason to delete it.
Connect. Explore. Keep watching.
Start exploring your own data.
Use Chartbrew Cloud or host it yourself.
