By Jonah Feld
We are not epidemiologists. We are experts in data visualization, and we put that expertise to use to better understand the COVID-19 crisis.
The New York Times collected this dataset, which it “made available to the public in response to requests from researchers, scientists and government officials who would like access to the data to better understand the outbreak.”
The following eight interactive visualizations combine this COVID-19 data with US Census data, allowing users to filter and explore critical information at a state and county level.
We took raw numbers for cumulative cases and cumulative deaths by county and by date and turned it into something that allows the user to explore to better understand the meaning of raw data.
We also created measurements, like ratios, rates of change, percentages, etc., that add context. And we added filters to let users narrow the scope to create their own visuals that answer the most important question in data interpretation: “Compared to what?”
Through these graphs, we hope to visually communicate how serious this disease is, as well as its overall impact to date and the effectiveness of efforts so far to flatten the curve.
We hope these graphs help us all better understand this difficult situation.
Descriptions of the types of graphs:
Treemap: A treemap visually represent parts of a whole. Contrasted with a map, the states and counties are proportionate to the desired value, not land area.
Animated Scatter: An animated scatterplot is a fantastic visualization when a line chart won’t do. Click on a single dot to trace progression over time.
Racing Chart: A racing chart is like a flipbook of bar charts over time. For a single measurement, it communicates changes in comparative rank, differences, and scale in an easily digestible clip.
Historic Table: This sortable table shows most metrics through the point in time indicated in the date picker. Right-click on a state to drill down to county, or skip to all counties nationally.
Daily Chart: These charts show daily activity (rather than cumulative activity as of each date). The vertical orientation encourages comparisons of daily measures over a shared x axis for dates.
Log Chart: A logarithmic scale is best for representing exponential functions. A steady slope of a line represents a fixed rate of exponential growth, and the horizontal gridlines indicate a relative change in magnitude, typically 10x.
Since Inception: This chart replaces the dates on the x axis with the number of days since reaching a common starting point: either 100 cases or 10 deaths. By aligning start points, the differences in rate of growth are more easily observed.
Small Multiples: Small multiples, popularized by visualization guru Edward Tufte, are a matrix of similar graphs using the same scale and axes, allowing them to be easily compared.
Jonah Feld is a Director of Product Development at Acronym. He specializes in data visualization and data integration for Keyword Objects, Acronym’s proprietary Enterprise Keyword Management platform for professional search engine marketers. He has 17 years of experience in SEO/SEM, with a heavy focus on analytics and reporting, having worked at agencies and in consultative roles developing business intelligence solutions.