On the Cover: Violin Plot
Zhongmin Cui · Educational Measurement Issues and Practice · 2020
Violin plots are less known than box plots, although they are cousins in that they show similar information. In addition to median and interquartile range, a violin plot shows the distribution of the data with a rotated smoothing curve. The distribution reveals additional details on the data (e.g., the number and location of peaks). A violin plot can be either elegant like a real violin or rough like the nose of a sawfish––it depends on the shape of the data and the smoothing function. Used properly, violin plots are appealing and insightful. This issue's cover features The Musical Relative of the Boxplot– The Violin Plot by Aron Fink from Goethe-University Frankfurt am Main in Germany. “The graphic shows a violin plot of the bias of the ability estimates for high ability students resulting from a computerized adaptive test using four different algorithms,” Fink described. “The middle part of each violin shows a classical box-whisker-plot,” the German researcher continued. “The shape of the violin displays the distribution of the data.” Fink created the graphic using the ggplot2 package in R. This package can be used to create many elegant data visualizations, violin plots included. Interested readers may refer to https://cran.r-project.org/web/packages/ggplot2/index.html for details about the ggplot2 package. To sum up, violin plots provide an advanced visualization of the distributions in a compact and attractive format that has several advantages over common approaches like classical box or density plots. They provide practitioners and educational researchers with a novel graphical tool which can easily be created using standard software like ggplot2. If you are interested in learning more about this informative data visualization, please contact Aron Fink ([email protected]). If you have any questions or suggestions on the cover in general, please email Zhongmin Cui ([email protected]).