A violin plot is a chart that shows the distribution of a numeric variable for one or more groups. It looks like a box plot wrapped in a mirrored density curve. The curve shows where data points are dense, and the box inside shows the median and quartiles.

It is used to compare the shape, spread, and centre of data across groups. Researchers use it when the mean alone is not enough, for example when data is skewed, has outliers, or has more than one peak that a bar chart would hide.

Wider parts of the violin mean more data points fall at that value. The white box shows the interquartile range, the dot or line marks the median, and the whiskers show the typical range. Long tails signal skew, and two bulges signal two clusters.

A box plot only shows five summary numbers: the minimum, Q1, median, Q3, and maximum. A violin plot adds the full density curve on top, so you can see the actual shape of the distribution, including skew and extra peaks that a box plot misses.

Aim for at least 10 points per group, and ideally 20 or more. With fewer points the density curve is unstable and can suggest patterns that are not real. For tiny samples a dot plot or strip plot is safer.

Width shows density. A wide section means many data points share that value, and a narrow section means few. The widest point is usually near the median for symmetric data, or shifted up or down for skewed data.

Yes. If a group has two clusters of values, the violin shows two bulges. This is called a bimodal shape and is one of the biggest reasons to choose a violin plot over a bar chart or a simple box plot.

Bandwidth controls how smooth the density curve is. A large bandwidth gives a smooth, simple shape; a small one shows more bumps and detail. This tool starts from the Silverman rule and lets you fine-tune it with a slider.

If the violins are skewed, report the median because it is not pulled by extreme values. If they are roughly symmetric, the mean and median will be close and either is fine. Many journals prefer the median with the interquartile range.

For two groups use a t-test or Mann-Whitney U. For three or more groups use one-way ANOVA or Kruskal-Wallis. Use the non-parametric options (Mann-Whitney, Kruskal-Wallis) when the data is skewed. This tool picks a sensible test on Auto.

They are close cousins. A bean plot also mirrors a density curve, but it usually shows individual data lines and a mean line instead of a box. A violin plot shows a smooth curve with a box inside. Both display distribution shape.

Use this online violin plot maker. Paste your numbers or upload a CSV, and the tool draws the chart, adds a box plot, runs a test, and exports an image. You do not need to install software or write any code.

Yes. Open the Plot Options and set orientation to horizontal. The value axis moves to the bottom and the group names move to the side, which is handy when group names are long.

Set a large export width and height in the Plot Options for a high-resolution image, then click Download PNG. PNG keeps sharp lines and text, which is what most journals want. JPEG and WebP are also available.

The density curve is an estimate, so it can spill a little past the real minimum and maximum. This tool forces each violin to taper to a crisp point, but a small overshoot near the tails is normal and expected.

Yes. Turn on Show data points in the Plot Options to add a light jitter of raw points over each violin. This is useful for small samples so readers can see the actual observations.

Pick the colour-blind safe scheme for accessibility, or match your journal style. You can also set a custom colour for each group. Keep fills semi-transparent so the box plot stays visible inside.

No. All calculations run in your browser on your own device. Nothing is uploaded to a server, so your data stays private. You can even save the page and use it offline.

Yes. Two groups work perfectly. The tool will offer a t-test or Mann-Whitney U instead of ANOVA and draw a bracket with a star when the difference is significant.

Describe the plot in your methods, name the test you used, and note the software. You can write, for example, "Violin plots were produced with the StatsUnlock Violin Plot Maker (statsunlock.com)."