Charts

Box Plot Generator

Quartiles, whiskers and outliers for one group or many, drawn as a downloadable SVG box plot.

Loading the tool…

Processing happens locally in your browser. What you paste or load is processed by this page and is not uploaded to a server. Nothing is stored unless you use a control that says it stores something, and you can clear anything this site has kept from the privacy page.

How to use this tool

  1. Paste a column of numbers for a single box, or a group name and a value on each line to compare several.
  2. Select Draw the box plot.
  3. Read the box first — it holds the middle half of the values — and then the whiskers, which show how far the rest reaches.
  4. Look at the circles beyond the whiskers. Those are the points an average would have hidden from you.

What box plot generator does

A box plot answers a question a bar chart of averages cannot: how spread out is this, and is the average telling the truth? Two groups with identical means can have completely different distributions, and the chart that shows only the mean is the reason nobody notices until the tail becomes somebody’s incident.

Paste a column of numbers for one box, or a group name and a value on each line to compare several side by side. The box spans the first to the third quartile with the median marked inside it, the whiskers reach the furthest point within 1.5 times the interquartile range, and anything beyond that is drawn individually as an outlier. Switch the whiskers to the true minimum and maximum when you would rather not label anything an outlier.

Frequently asked questions

The box spans the first quartile to the third — the middle half of the values — with the median as the line inside it. So half of everything sits inside the box whatever the group size, which is what makes several boxes side by side genuinely comparable even when the groups have very different counts.

By Tukey’s convention: anything further than 1.5 times the interquartile range beyond a quartile. It is a convention, not a test — it does not mean a point is wrong or should be removed. It means the point is unusual relative to the middle of this particular distribution, which is a reason to look at it, not to delete it.

A histogram shows one distribution in full detail, including whether it has two peaks — something a box plot conceals completely. A box plot compresses each distribution to five numbers, which is what lets you put eight of them side by side. Use the histogram to understand one group and the box plot to compare many.

At least five values per group, which is the minimum for quartiles to mean anything at all; below about twenty the quartiles are still jumpy. It is far more forgiving than a histogram at small sizes, because a quartile moves much less than a bin count when one value changes.