Raincloud plots: a multi-platform tool for robust data visualization

Across scientific disciplines, there is a rapidly growing recognition of the need for more statistically robust, transparent approaches to data visualization. Complementary this, many scientists have called plotting tools that accurately and transparently convey key aspects statistical effects raw with minimal distortion. Previously common approaches, such as conditional mean or median barplots together error-bars been criticized distorting effect size, hiding underlying patterns in data, obscuring assumptions upon which most commonly used tests are based. Here we describe visualization approach overcomes these issues, providing maximal information while preserving desired ‘inference at glance’ nature other similar devices. These “raincloud plots” can visualize probability density, summary statistics median, mean, relevant confidence intervals an appealing flexible format redundancy. In this tutorial paper, provide basic demonstrations strength raincloud plots outline potential modifications their optimal use, open-source code streamlined implementation R, Python Matlab (https://github.com/RainCloudPlots/RainCloudPlots). Readers investigate R tutorials interactively browser using Binder by Project Jupyter.