COVID Gowns: Breaking Scales with an Unknown End by Jackie Yu, Neil Chulpongsatorn, Sarah Storteboom, Sheelagh Carpendale, Soren Knudsen, Tatiana Losev
COVID Gowns is an interactive online visualization of Canadian COVID-19 cases. A common horizontal timeline or horizontal bar chart is limited by how much data can be included because the data will eventually become visually compressed and distorted. Therefore, we represent time vertically along gown-like shapes, which allows data to be continually added without visually warping case trends. This feature is useful as the pandemic continues in waves of case counts over time.
Each Gown represents cases from a province showing daily and cumulative COVID cases. The most recent date is at the top of each gown. Viewers can interact with the visualization by scrolling and compare case counts across provinces and dates. By hovering, COVID Gowns displays the exact date, the number of daily cases, cumulative cases. Additionally, by clicking on these numeric displays, the Gowns (Canadian provinces) can be realigned based on these numbers to compare trends between dates and regions.
On each Gown, time is displayed by day on the y-axis, while daily new cases and cumulative cases are on the x-axis shown on the left and right sides respectively. The design uses a hybrid logarithmic base 2 scale by layering semi-transparent linear scale graphs on top of each other. Each graph is sized to fill the column and ends when case numbers double. The visualization combines details and an overview of case counts by overlapping graphs and visually distinct scale breaks. The scale breaks show interval lengths between doubling cumulative cases. The breaks are visible by layering distinct graphs with different colours and opacities to ensure that the ends of the graphs are still visible. This change is less evident in the bar charts of COVID Gowns where there is more variability in the data.
Hovering over a value shows a dotted line that links to another province where the same value occurs. Clicking on a given value repositions the Gowns to the date of the same value in other province Gowns. This interaction provides flexibility in how the viewer wishes to align the Gowns, along with a view of the data before and after that alignment.
The vertical timelines are compact and flexible. Scrolling downwards gives enough space to monitor local outbreaks on mobile devices. On a desktop computer, the horizontal space allows several provinces (Gowns) to be repositioned to compare trends, or COVID waves, across regions — this feature also shows many valuable zeros. COVID Gowns offer flexibility in configuring comparisons between regions and monitoring possible correlations between policy interventions, or other factors in case trends for future research and public health interventions.