OriginPro: Beyond Traditional Box Plots, How to Draw a Beautiful Raincloud Plot

Introduction

#

A raincloud plot is a visualization method that combines the shape of a data distribution, individual observations, and key statistical summaries. It is formed from a rotated half violin plot and usually consists of a half violin plot (the “cloud”), jittered or binned raw data points (the “rain”), and optional summary information such as a box plot, mean, or median. Its purpose is to overcome the information loss and misleading impressions that traditional bar charts or mean-with-error-bar charts can create, using a more intuitive, modular, and statistically robust display. Allen and colleagues systematically explained this approach. Its central value is that it shows both the raw data and the distribution shape, reducing the viewer’s need to guess and removing visual redundancy.

A raincloud plot is not universal, though. When data points are extremely dense, especially in repeated-measures data, or when there are too many groups, the point cloud and density layer can become crowded and harder to read. When you only need to show simple counts or proportions, a bar chart or stacked bar chart may be enough.

Code implementations (Python/Matlab/R): RainCloudPlots open-source repository (maintained by the authors).

Data Requirements

#

The data must be numeric.

Column Arrangement and Labels

#

• Minimum data volume: at least one numeric data column can generate the chart, though side-by-side comparison across multiple columns is more common.
• Group naming: column labels such as long names or comments are used for axis ticks and legends. You can maintain them in the column-label rows before plotting, and they will be inherited automatically by the chart.

OriginPro Plotting Tutorial

#

A raincloud plot is essentially a rotated half violin plot + individual data points + optional statistical summary. In OriginPro, there are two ways to build it.

Operation tip (GUI mode): Origin 2025b added “Stats GUI Mode,” which can be switched under Preferences → GUI Mode. This article describes menu paths in the default mode. If you cannot find a menu item, switch back to the default mode first.

Draw the Basic Chart

#

Method 1: Build Manually (Half Violin Plot + Rotation)

#

1. Select data: in the worksheet, select all Y data columns to be plotted.
2. Create a half violin plot: use Plot → Statistical → Half Violin from the menu bar.
3. Swap axes: click Exchange X-Y Axes on the graph toolbar at the right, or use Graph → Exchange X-Y Axes from the menu to rotate the plot into a horizontal display.

Method 2: Use the Raincloud Plot Template Directly

#

1. Use Tools → Template Center to open the online template library, search for “Raincloud”, and install the template. The Template Center downloads the online template and provides it under “Extended Templates.”
2. Return to the data sheet and choose data columns → Plot → Extended Templates → Raincloud Plot.

Chart Styling

#

After generating the basic chart, you usually need fine-grained adjustments to meet publication standards. Double-click any chart element, such as a data point or axis, to open the Plot Details dialog.

Adjust the “Cloud” and “Rain”

#

1. “Cloud” (density plot): select the corresponding data plot in the left panel, then switch to the Distribution tab.

   • Fill and color: in the Fill section, configure a visually pleasing color set by Fill to Curve.
   • Size and width: under Size, I recommend checking Width so the maximum widths of all groups are visually comparable. Adjust Scale to Max (%) to control the overall proportion.

2. “Rain” (raw data points)

   • Style: on the Symbol tab, set Shape to Circle and Fill to Open. For readability, set Size (Z) to around 3 pt and Edge Thickness (T) to 0.5–1 pt.
   • Distribution: if the data points are too sparse, check Binning on the Data tab and enter a larger value.

Configure the Box Plot and Statistical Summary

#

1. Box: switch to the Box tab and choose Regular Box under Style. Adjust Box Width (%) to match the overall visual design. On the Pattern tab, you can change the border and fill color.
2. Whiskers: on the Box tab, the whisker Range field defines how whiskers are calculated. A standard raincloud plot usually uses Min-Max. Whisker line style can be changed under the line tab for whiskers, including line style, color, and width.
3. Percentiles: on the Percentile tab, you can check and customize marker styles for Max, Min, Mean, and similar markers. Set max/min markers to | and the mean to 〇.

Configure Axes and Layers

#

1. Open settings: double-click any axis to open the Axes settings dialog.

2. Axis lines and ticks

   • In the left panel, hold Shift and click Bottom, Left, Top, and Right in order to select all axes at once.
   • Switch to the Line and Ticks tab and set the Style of both Major Ticks and Minor Ticks to In. Adjust their length and thickness as needed.

3. Grid lines

   • In the left panel, select the Vertical axis alone.
   • Switch to the Grids tab, check Show for major and minor grid lines, and adjust their style, such as dashed lines, and thickness to improve readability.

4. Layer frame

   • After closing the settings dialog, click the graph. In the floating toolbar, click Layer Frame (second button in the first row) to add a complete border to the chart.

Reading a Raincloud Plot: Three Core Elements

#

To understand the structure of a raincloud plot more intuitively, we can break down its three core components, the “cloud,” the “rain,” and the statistical summary, and explain how to read each one.

1. “Cloud”: Probability Density of the Data Distribution

#

The “cloud” is the signature element of a raincloud plot. It is essentially a half violin plot. The width of the curve reflects the relative density of observations at that value: wider regions indicate more concentrated data, while narrower regions indicate sparser data. This combination of “shape + amount” reveals details such as skewness, long tails, and multiple peaks.

2. “Rain”: Individual Display of Raw Data

#

The “rain” consists of jittered or binned raw data points, which ensures visual transparency. You can directly see where each observation lies, how dispersed the observations are, and whether clusters or outliers exist. For large sample sizes, reduce point size or strengthen binning/jittering to avoid overplotting.

3. Summary Statistics: Condensed Key Indicators

#

A raincloud plot can overlay summary information such as median/mean points, boxes, and whiskers (commonly minimum to maximum). This provides reproducible numerical anchors while still showing the full distribution. If the box blocks the density layer, adjust transparency or shift the box slightly. Specific components and meanings can be checked against Origin’s “Violin with Quartile / Box” documentation.

References

#

Allen, M., Poggiali, D., Whitaker, K., Marshall, T. R., van Langen, J., & Kievit, R. A. (2021). Raincloud plots: a multi-platform tool for robust data visualization. Wellcome Open Research, 4, 63. (Version 2, published 21 January 2021). https://doi.org/10.12688/wellcomeopenres.15191.2