How to Make a Violin Plot in GraphPad Prism

How to create, format and read violin plots in GraphPad Prism, from data entry to smoothing and quartile lines.

GraphPad Prism has drawn violin plots since Prism 8. You enter your data in a Column table (or a Grouped or XY table with replicates), choose a violin plot when you create the graph, and then set smoothing and median and quartile lines in the Format Graph dialog. This guide covers each step, the options that change the shape, and how to read the result.

Menu names and dialog layouts shift between Prism versions, so the steps below describe where things are in general terms. If you do not have Prism to hand, you can make a violin plot online by pasting the same columns.

Which Prism versions support violin plots

Violin plots arrived with Prism 8 and are available in all later versions. Early Prism 8 releases always stopped each violin at the smallest and largest data value. A later 8.x update added the choice to let the density curve extend beyond them. If an option described here is missing, check that your copy is up to date.

Step 1: enter the data

The usual setup is a Column data table with one column per group. Each row is one observation, and columns can have different lengths.

ControlDrug ADrug B
4.15.63.2
3.86.07.9
4.55.23.5
………

Prism can also draw violins from replicates entered side by side in a Grouped or XY table. Recent versions can graph straight from a Multiple Variables table without restructuring it first. For two grouping factors, such as treatment and sex, a Grouped table is usually the natural fit.

A violin needs enough points per group to estimate a shape. With very small groups, often under ten or so values, the outline mostly reflects the smoothing setting rather than your data. Show the points in that case.

Step 2: create the graph

Create a new graph of the data table. In the graph type gallery for column data, Prism groups box-and-whisker and violin plots together. Choose the violin option. You can change the details afterwards in Format Graph.

Step 3: format the violins

Open the Format Graph dialog, usually by double-clicking a violin. The violin settings live there.

Smoothing

Prism draws each violin from a kernel density estimate. It offers three smoothing levels: light, medium and heavy. GraphPad’s own guide describes it this way: light smoothing shows more detail of the distribution, heavy smoothing gives a better sense of its overall shape.

A practical approach is to turn on the individual points, flick through the three levels, and keep the one where the outline follows the points without wiggling around single values. This is the same bandwidth trade-off covered in our violin plot bandwidth guide. Prism just presents it as three presets.

Median and quartile lines

Prism draws lines across the violin at the median and at the 25th and 75th percentiles. You can style the median line and the two quartile lines separately: pattern (solid, dashed, dotted), colour and thickness. A common style is a solid, thicker median and thinner dashed quartiles. That makes the median stand out without hiding the shape.

To style one group differently from the rest, select that violin on the graph and change it there, or in the Graph Inspector in newer versions, instead of applying the change to all data sets.

Truncated or extended ends

By default, Prism stops each violin at the lowest and highest observed value, so the ends are flat. A kernel density estimate naturally extends a little past the data. Prism trims it because a violin is meant to show the data you have. Versions that offer the truncate choice let you switch it off to draw rounded tails. Keep truncation on for values with hard limits, such as percentages or counts. Extended tails there can suggest impossible values.

Fill and points

You can fill the violin with a colour and overlay the individual data points. If the points are the same colour as the violin, the ones at the very top and bottom can disappear under the flat caps. Use a contrasting colour or a slightly larger symbol. For more on this layout, see our guide to a violin plot with points.

A note on logarithmic axes

Prism calculates the violin from the original values, not from the axis scale. On a log Y axis this causes two problems. The shape no longer matches how the points are spread, and the density can reach below zero, which a log axis cannot show. GraphPad recommends against combining the two unless you know exactly what the graph means.

The cleaner route for skewed data such as concentrations is to log-transform the values first, using Prism’s Transform analysis, then graph the transformed column on a linear axis. Add tick labels that show the original units if you need them.

Reading a Prism violin plot

  • Width at any height shows how many values sit near that level. Wider means more common.
  • The median line marks the middle value. The quartile lines enclose the middle half of the data.
  • Two bulges suggest two subgroups, such as responders and non-responders. A box-and-whisker plot would hide this.
  • A long thin tail on one side means skew. The median and the mean will differ.
  • The flat ends are the minimum and maximum when truncation is on, not features of the distribution.

Be careful comparing the widths of different violins. Depending on settings, violins may be scaled independently, so a wider violin does not always mean a bigger group. Report sample sizes in the legend or below the axis. For a full walkthrough, see how to read a violin plot.

Violin or box-and-whisker in Prism?

Both graph types come from the same data table, so you can make both and compare. GraphPad’s documentation suggests violins as an alternative to box-and-whisker plots because they show the frequency distribution. A box plot is still easier for readers who only need the median and spread, and it holds up better with small groups. Our violin plot vs box plot article covers how to choose.

Statistics are separate from the graph

A violin plot is a description, not a test. To compare groups, run the analysis you need from the same data table, such as a t test, one-way ANOVA or a nonparametric test. Prism can then add significance markers to the graph. Choose the test based on your study design, not on how the violins look.

Alternatives if you do not have Prism

Prism is paid software. For one figure, our free violin plot maker accepts pasted columns in the same layout as a Prism Column table and exports an image. For scripted, reproducible figures, R and Python both have mature violin functions. See the violin plot in R and violin plot in Python guides. The violin plot guide covers the concepts that apply in every tool.

FAQ

Which version of Prism has violin plots?

Prism 8 and later. Later 8.x updates and newer versions add more control over details such as whether the ends are truncated.

What data table do I need?

A Column table with one column per group is the simplest. Grouped and XY tables with replicates also work, and recent versions can graph from a Multiple Variables table.

Why are my violins flat at the top and bottom?

Prism truncates the density at the highest and lowest data values, so the flat ends mark the maximum and minimum.

How do I change the smoothing?

In the Format Graph dialog, choose light, medium or heavy smoothing. Turn on the data points to judge which level fits.

Adrian Lee
Adrian Lee

Adrian Lee is the Data Visualization Specialist and Statistical Analyst at ViolinPlot. Based in Boston, Massachusetts, Adrian utilizes a robust background in statistics and data science to create compelling and insightful visualizations that help interpret complex data.

With extensive experience in data analysis and visualization techniques, Adrian specializes in developing violin plots and other advanced graphical representations to convey statistical information effectively. Adrian’s expertise includes statistical modeling, data interpretation, and the creation of interactive visualizations.

At ViolinPlot, Adrian is responsible for transforming raw data into visual narratives that are both informative and engaging.

By employing state-of-the-art visualization tools and statistical methods, Adrian ensures that the visual representations are accurate, clear, and valuable for decision-making processes.

Adrian’s dedication to enhancing data comprehension through visual tools has established ViolinPlot. As a leading resource for data visualization and statistical analysis.

Articles: 89

Ready to Unlock Deeper Data Insights?

Transform complex datasets into clear, interactive visualizations with Violin Plot. Start exploring data distributions in a way that empowers insight and drives smarter decisions.

LEARN MOre
100% satisfaction guaranteed
McAfee SECURE
ssl-secured
verisign-secured
BEST-PRICE-GUARANTEED-free-trust-badge
guaranteed-safe-checkout-1