Visualizing compositional data on the Lexis surface

Background: The Lexis surface plot is an established visualization tool in demography. Its present utility, however, is limited to the domain of one-dimensional magnitudes such as rates and counts. Visualizing proportions among three or more groups on a period-age grid is an unsolved problem. Objective: We seek to extend the Lexis surface plot to the domain of compositional data. Methods: We propose four techniques for visualizing group compositions on a period-age grid. To demonstrate the techniques we use data on age-specific cause-of-death compositions in France from 1925 to 1999. We compare the visualizations for compliance with multiple desired criteria. Results: Compositional data can effectively be visualized on the Lexis surface. A key feature of the classical Lexis surface plot – to show age, period, and cohort patterns – is retained in the domain of compositions. The optimal choice among the four proposed techniques depends primarily on the number of groups making up the composition and whether or not the plot should be readable by people with impaired colour vision. Contribution: We introduce techniques for visualizing compositional data on a period-age grid to the field of demography and demonstrate the usefulness of the techniques by performing an exploratory analysis of age-specific French cause-of-death patterns across the 20th century. We identify strengths and weaknesses of the four proposed techniques. We contribute a technique to construct the ternary-balance colour scheme from within a per- ceptually uniform colour space. Comments: A full-colour representation is key to understanding the paper. Therefore, we recommend that you read it on screen or print a colour version.