Exploring many views on one pandemic
By: Megan Danielson
You’ve no doubt seen many COVID-19 charts, maps, and data visualizations shared across media outlets over the past few weeks. These maps can be powerful tools for understanding information and making data-driven decisions. However, the fidelity of these visualizations is limited by the underlying data — which varies by source, region, and resolution — and that data’s ability to shed light on a specific facet of the epidemic.
The fact is, there’s no one-true data source and no one-true way to visualize the data.
We’ve been working closely with researchers and public officials to explore and support using different data to help understand multiple aspects of this outbreak. The maps below represent some of the varying approaches and possibilities of using data to gain insight into the spread of and response to COVID-19.
NextStrain
NextStrain builds tools to visualize the evolution and spread of the virus in real-time specifically for virologists, epidemiologists, public health officials, and community scientists.
Infodemics Observatory
Infodemics Observatory is visualizing the digital conversation and information spread around the COVID-19 outbreak. Using machine learning techniques, it analyzes over 100 million public messages to understand dimensions of collective sentiment and psychology, news reliability, and the extent to which social bots are influencing discourse with unreliable news.
Datawrapper
Datawrapper is leading by example in the data visualization community, creating more than 20 publicly available charts, maps, and tables that show the latest numbers on the Coronavirus in a responsible and informative way.
USA Today
Many visualizations, including USA Today’s COVID-19 tracker, use data resources collected by Johns Hopkins University (JHU) to track the spread of the disease in the US and worldwide. The JHU data is curated from a collection of official reports posted by the WHO, the CDC, ECDC, NHC, and DYX. The map visualizes the cumulative spread of the virus at varying geographic resolutions and is often accompanied by supporting information on COVID-19 case status including the number of new cases, number of recovered, and number of deaths.
COVID-19 (Coronavirus) Data Resource Hub
The Tableau Foundation, in partnership with Mapbox, has made it easy to integrate additional data with the case reporting data from Johns Hopkins University in order to help organizations contextualize the spread of the disease.
HealthMap/Boston Children’s Hospital
The HealthMap collects publicly available information on confirmed cases during the ongoing COVID-19 pandemic. This data is granular and includes contextual information that may be useful to researchers such as age, sex, geographic information, history and time of travel where available. Data is available here. We are currently working with HealthMap to redesign this visualization in order to enhance users' ability to gain insights from their data.
HealthMaps other visualization offers high-frequency updates derived from the news articles and official reports. The research team uses web scrapers/crawlers to target publicly available articles and reports that contain infectious disease keywords. The system is fine-tuned based on a weighing of sensitivity vs specificity, which is set at the infrastructure level (rather than for each alert or data feed, etc). Once ingested, the data undergoes a process of automated classification using NLP/ML algorithms, which assigns disease, location, species, case count, and “place category” tags.
EpiRisk
EpiRisk is unique in that it allows for a quick estimate of the probability of exporting infected individuals from sites affected by a disease outbreak to other areas in the world.
Data Localization
Nikkei Asian Review
Using data reported by the WHO and China’s National Health Commission, Nikkei Asian Review created a visualization that allows users to explore the scale of the outbreak in China at the provincial level, over time.
San Francisco Chronicle
Visualizing confirmed cases of COVID-19 by county for all of California, the SF Chronicle provides a highly localized view for one state. The map is updated daily with data pulled directly from the reports by the U.S. Centers for Disease Control and Prevention, the California Department of Public Health, and county public health departments.
Mask Guide 口罩指南 by Che-Lin Chan and Chi-Yung Yung
Mask guide is an application that helps individuals in Taiwan locate stores that are selling face masks. (Note: the WHO and the CDC are not currently recommending the use of masks to prevent the spread of COVID-19 unless you are taking care of an individual with suspected COVID-19 infection. WHO guidelines for masks.)
Japan COVID-19 Tracker
The Japan COVID-19 tracker pulls in data from the Japanese Ministry of Health. The solution helps Japanese residents understand the current conditions of the outbreak and allows users to directly download their source data.
We’ll continue adding to this post as new visualizations and perspectives emerge. If you see more notable examples, share them with us on twitter by tagging @mapbox.
The Mapbox Community team (as well as many other people across Mapbox) is working closely with many organizations to support data visualization and public health projects. If you are working on a new mapping or data project that needs support, reach out to me.
Megan Danielson - Community Program Manager - Mapbox | LinkedIn
The Mapbox Community Team provides tools and resources to individuals and organizations using maps for positive impact. Learn more about Mapbox Community and get in touch with our team.
Maps feature data from Mapbox and OpenStreetMap and their data partners.
Notable maps visualizing COVID-19 and surrounding impacts was originally published in Points of interest on Medium, where people are continuing the conversation by highlighting and responding to this story.