Hammock plots: visualizing categorical and numerical variables

I discuss the hammock plot for visualizing categorical or mixed categorical/numeric data. Hammock plots can be viewed as a generalization of parallel coordinate plots where the lines are replaced by boxes (or plotting elements) and the width of the boxes is proportional to the number of observations they represent. The paper also introduces a modification to the hammock plot to avoid what Hoffman et al. termed the reverse line width illusion. Further, I give an historical overview over hammock-type plots such as common angle, GPCP, parsets, and alluvial plots and discuss the type of boxes used to connect adjacent variables. Supplemental materials are available online.

本文探讨了用于可视化分类数据或分类/数值混合数据的吊床图（hammock plot）。吊床图可视为平行坐标图（parallel coordinate plots）的推广形式，其将平行坐标图中原有的线条替换为箱体（或绘图元素），且箱体宽度与其所代表的观测值数量呈正比关系。本文还提出了一种针对吊床图的改进方案，以规避Hoffman等人所称的反向线宽错觉。此外，本文对吊床类绘图方法（包括通用角度图、GPCP、parsets以及冲积图（alluvial plots））开展了历史沿革梳理，并探讨了用于连接相邻变量的箱体类型。补充材料可在线获取。