Topics: Coronavirus | AI | Data Science | Deep Learning | Machine Learning | Python | R | Statistics

KDnuggets Home » News » 2020 » Apr » Tutorials, Overviews » Understanding the COVID-19 Pandemic Using Interactive Visualizations ( 20:n18 )

Understanding the COVID-19 Pandemic Using Interactive Visualizations


Interactive visualizations are an effective method for understanding the COVID-19 pandemic. This article presents a repository filled with just such insightful interactions.



Early last month, as the coronavirus further spread, disrupting (and, unfortunately, ending) life for so many, and as the exponential growth continued to manifest, I wanted to find a simple and clear set of visualizations to track the daily and cumulative confirmed cases and deaths.

As the month of March trudged along, and as COVID's wrath was being felt in Europe and increasingly in North America, I shared the work of John Burn-Murdoch with the readers of KDnuggets, on the chance they had not been making use of it to that point. John works on data visualization (or, as his Twitter bio says, "Stories, stats & scatterplots") for the Financial Times. He has recently garnered a following and appreciation for his COVID-19 visualizations, which are exemplars of how quality data visualizations can clearly and succinctly tell a story. John contributes to the daily-updated Financial Times article tracking COVID spread, made available for free to non-subscribers in this weird coronavirus era.

As more data is collected and becomes publicly available, John's work has expanded to include the analysis of different COVID pandemic angles. So, too, have the opportunities grown for others to expand their analysis in the same way. This increased data, and lengthened pandemic time period, also allows some opportunity for interactive visualizations of COVID data. The Coronavirus Disease (COVID-19) data repository at Our World In Data (OWID), put together by Max Roser, Hannah Ritchie, Esteban Ortiz-Ospina and Joe Hasell, is a great example of this.

First, OWID does not focus solely on COVID. They have been around for a while and are interested in doing public good. From their about page:

 

The goal of our work is to make the knowledge on the big problems accessible and understandable. As we say on our homepage, Our World in Data is about Research and data to make progress against the world’s largest problems.

 

OK, so what makes their COVID data visualization site worthy of mention? The interactive tool below should whet your appetite.


Total confirmed COVID-19 deaths in Nordic countries.

 

Given the interest in Sweden's distinctive approach to dealing with the pandemic, for demonstration purposes we will use the above chart which compares and contrasts the total confirmed deaths from the disease from the Nordic countries.

Note that this is a comparison of the absolute numbers. Don't go out of your way to comment on the lack of per capita normalization for population; we all get it already. I am of the opinion that there are reasons to use the absolute numbers, and there are reasons to used normalized per million numbers, though many people seem to have a very hardened opinion as to which is the "correct" way. I won't discuss the matter beyond saying that population-based normalizing versus not provide different insights into the data, both of which can be useful. So long as you know which is being used, and any caveats in their use, this really shouldn't be an issue. There is a deaths per million graphic below.

Back to the visualization tool above: You will see the absolute numbers of confirmed deaths in each of the Nordic countries. It is presented in a y-axis linear scale, but if you want to see what this looks like in a log scale, click the word "linear" in the top left corner of the visualization, and voila. You will also see a slider along the x-axis, should you want to clip the dates shown in the visualization (I left the defaults). The additional map, data, and sources tabs along the bottom will change what the tool is presenting likely consistent with your intuition of what these tabs should present.

As promised, here is an interactive map of the total confirmed global deaths per million.


Total confirmed COVID-19 deaths per million globally.

 

Highlighting any country will provide the confirmed deaths per million, while clicking on a country will switch to a chart presenting the change in confirmed deaths over time. The slider on this graphic won't clip any dates; it's a time machine to show the per million count on a given historical date.

A this point in the pandemic, what stands out to me as the most dramatic insights are those which are date-related. Take a look at the confirmed deaths for the United States below:


Total confirmed COVID-19 deaths in the United States of America (Feb 28 to today).

 

Note that it is not my intention to single out the US (or Sweden), as each country is dealing with their own degrees of pandemic crisis at the moment, but both of the above charts are useful for demonstration.

What is incredible is that, 2 months ago at the time of writing (Feb 29), the United States did not have a single confirmed COVID-19 death. One month later (Mar 29) there were 2191 confirmed deaths. As of today (Apr 29) there were 58355 confirmed deaths. The growth is staggering, and seeing it visualized makes it even more so. Presenting this data with a linear y-axis (as opposed to log) makes this all the more striking, at least in my view.

Beyond confirmed death charts, OWID's COVID-19 site allows for the exploration of various types of interactive visualizations (maps, bar charts) on a wide array of data (confirmed cases, testing, healthcare capacity). Check out the site for yourself and see what might be of interest to you. There are myriad approaches to helping understand the pandemic through the interactive visualizations OWID has made available.

Hopefully as we continue to get a handle on the spread of the virus and its disease, we will be able to glean positive trends and insights from this same data. You can access OWID's raw COVID-19 data directly here. You can read more about their data here.

 
Related:


Sign Up

By subscribing you accept KDnuggets Privacy Policy