Charts

Violin Plot

A hybrid plot displaying density curves alongside summary elements.

What is a violin plot?

A violin plot is a hybrid between a box plot and a mirrored density plot and is used to visualize the underlying distribution of a dataset. The probability density of the data is smoothed using a gaussian kernel, where wider sections of the violin plot indicate where more data points are concentrated.

Traditionally, a box plot is layered on top of the density plot with the "middle dot" representing the median value, the "box" displaying the interquartile range (IQR) and the "whiskers" (thin line) showing either the range, Tukey's fences, or percentiles.

Graphmatik also gives you the option to overlay a spread plot instead of the more traditional boxplot. Here the center dot represents the mean ± error (i.e. range, SEM, SD, 95% CI).
WARNING: Violin plots showing the mean ± error are NOT as common as box plot overlays. So you should always describe what these chart elements represent to the reader.
Default vertical violin plot
A combination of a box plot and kernel density estimate
A vertical violin plot with Graphmatik's default styling applied.

When to use a violin plot

When combined with a distribution dot plot—"bar", "spread" and "box" plots can provide as rich an understanding of the underlying data distribution as violin plots including information about modality, skewness, and unlike violins can also show the number of observations.

Where violin plots excel, is working with larger datasets. When there is sufficient data points to reliably estimate the probability density and the sheer volume of data makes distribution dot plots appear visually cluttered and dense.

Make sure to check out these amazing tips on how to create beautiful column charts
Easily swap directions
Alternate between vertical or horizontal violin plots by switching types within the data workspace.
A horizontal violin plot showing four groups in descending order. The largest group (on top) is highlighted in purple.

Chart properties

PropDefaultDescription
central tendencymedian
median
The middle most value of a sorted set of numbers.
mean
The sum of a set of values divided by the number of values in the set.
whiskersrange
range median or mean
The difference between the highest and lowest values within a set.
1.5 * Interquartile range (1.5*IQR) median
A range representing Q1 - 1.5 * IQR and Q3 + 1.5 * IQR.
2.5 percentile - 97.5 percentile (2.5-97.5 %tile) median
The difference between the 2.5 percentile and the 97.5 percentile, representing the middle 95% of a set.
standard error of the mean (SEM) mean
How much the sample means vary from the population mean.
standard deviation (SD) mean
A measure of the variation of a set of values around their mean.
95% confidence interval (95% CI) mean
95% probability that the population parameter lies within this range.
sortnone
none
The dataset is arranged in insertion order.
ascending
The dataset is arranged from smallest to largest value.
descending
The dataset is arranged from largest to smallest value.

Box Plot

A classic plot for displaying a data distribution with quartiles.

Lollipop Chart

A simple, yet elegant alternative to the classic bar chart.