Violin Plot Builder

Create violin plots showing data distribution density alongside box plot summary statistics.

Tested tool guide Tested browser tools Checked August 16, 2026

What Violin Plot Builder does, with a checked example

A violin plot is a box plot wrapped in a mirrored density curve. Paste a column of numbers, or several columns for comparison, and this tool draws one violin per group: the box inside reports the median and quartiles, while the mirrored shape shows where the values actually bunch up, including separate peaks a box plot cannot reveal. The most common misreading is the width. Each violin is rescaled to the same area so groups can be compared side by side, so a wide violin means density, not more data.

Worked example

A concrete input and expected output from the current implementation.

Input

2, 3, 4, 5, 5, 6, 7, 8, 9, 12, 16

Expected output

One violin, widest around 6 to 7 where five values sit, narrowing through 8 and 9, then a long thin tail out to 16 and a small bulge near 2 to 3. The box inside spans the quartiles 4 to 9 with the median line at 6; the whiskers reach the true minimum 2 and maximum 16, and nothing plots beyond them. The shape reads right-skewed: the mean of 7 sits above the median of 6, matching the visible tail.

Eight of the eleven values lie between 2 and 9, so the density estimate bulges there and tapers sharply above. The isolated 12 and 16 pull the tail far past the median, which is the skew the box plot shows only as a longer upper whisker.

How the result is produced

Reading one violin

Each violin is a vertical density curve mirrored around a central axis, with a box plot drawn inside. The box marks the first and third quartiles, the line inside it the median, and the whiskers the rest of the data (or the conventional 1.5 interquartile-range fences). Values beyond the fences are outliers. The box is the summary; the curves are the detail.

Where the width comes from

At each value on the vertical axis, the curve's horizontal width is proportional to how many of your points sit near that value, smoothed into a continuous shape. The smoothing level sets the tradeoff between fine detail and broad form, and each violin's area is normalized so groups can be placed side by side. That normalization is why width carries no sample-size information.

Good uses

Compare one metric across groups, such as test scores by class or page load times by browser, to spot groups that are skewed, bimodal, or full of clumps that a plain average would hide.
Check whether a distribution is symmetric enough to summarize with the mean, or lopsided enough that the median is the honest summary, before reporting either number.
Replace a row of separate histograms with a single chart: paste several columns of numbers and compare their shapes, spread, and outliers at a glance.

Limits and checks

A wide violin does not mean more data. Because each violin's area is normalized, a group of 20 and a group of 2,000 can render equally wide; only the shape of the concentration differs. Keep the per-group sample sizes visible next to the chart.
Small samples make the curve mostly artifact. With a handful of points, the smoothed density shows spikes and flared ends that are properties of the smoothing, not of your data. Trust the box plot's quartiles for small groups and treat the density shape as suggestive.
The curve can spill past your data. The smoothed tails commonly extend beyond the lowest and highest values, so a flare is not evidence of hidden data; the whiskers mark the true range. The box's quartiles also follow one of several conventions, so they may differ slightly from the quartiles your spreadsheet reports.

Common questions

Why does my violin flare out beyond the highest value in my data?

The density curve is smoothed, and the smoothing reaches past the edge of the data. The flared tip is the curve's own width at the boundary, not a sign of unplotted values. The box plot is the reliable part for the actual range: when the flare and the whisker disagree, the whisker states the fact.

How many points do I need before the density shape means anything?

There is no hard cutoff, but below roughly twenty values the curve is dominated by smoothing artifact: spikes appear that vanish when you add a few more points. For small groups, read the box plot for the quartiles and the density only for rough shape, and state the sample size wherever the chart is shown.

References and verification

The example and behavioral notes were checked against the browser implementation. Standards and primary references below define the relevant format, formula, or platform behavior.