This column is written for those interested in learning about the world of Data for Good, where statistical analysis is dedicated to good causes that benefit our lives, our communities, and our world. If you would like to know more or have ideas for articles, contact David Corliss.

With a PhD in statistical astrophysics, David Corliss leads a data science team at Fiat Chrysler. He is the founder of Peace-Work, a volunteer cooperative of statisticians and data scientists providing analytic support for charitable groups and applying statistical methods in issue-driven advocacy.

From the very beginning of the COVID-19 pandemic, data and analysis have played a central role in understanding the crisis, informing health officials and hospitals, and driving data-driven decisions to address and mitigate the impact. It’s been all hands on deck for the D4G community, with everyone working together to make a difference.

COVID-19 Data Sources and Issues

Of course, analysis begins with the data. The Center for Systems Science and Engineering at The Johns Hopkins University has created a GitHub site to provide access to more than a dozen COVID-19 data sources, including a great wealth of international data. This site includes The Johns Hopkins database—the 2019 Novel Coronavirus COVID-19 (2019-nCoV) Data Repository—its daily updates summarized by geographic area from many places around the world. An excellent source for US data is from The New York Times, which has tabulated confirmed cases and deaths by county and date. These are just two of many data resources now online. Researchers investigating on a smaller geographic scale may find local data from reliable, official sources helpful.

In looking at the data, one concern discussed by subject matter experts is the under-reporting of cases, partly from incomplete testing and also due to COVID-19 often being asymptomatic and going undetected. These challenges have made fatalities the most reliable numbers. (Full disclosure: I am a statistician but not an epidemiologist.)

To track the spread of the pandemic, one method is to start with reported deaths for the raw data, divide by a reliably reported fatality rate (e.g., the value reported in The Lancet by Robert Verity et al.), and count back to the infection date. Plots of both cases and deaths often use a log scale on the vertical axis to reflect any interval of exponential growth.

So Many Opportunities

The American Statistical Association and its members have been quick to respond to this crisis in a number of ways. One is establishing an ASA community—COVID-19 Data, Statistics, Research, and Discussion—to support coronavirus/COVID-19 research. Researchers can share their work and resources, collaborate, and ask questions. Other ASA sections and communities have been active in supporting research or addressing challenges raised by the pandemic, such as expanded teaching online.

There are also many educational opportunities available. For example, the ASA has partnered with the University of Connecticut and others to sponsor a webinar series on data science in response to COVID-19.

All statisticians will appreciate the importance of incorporating subject matter expertise into any analysis. This is most critical in situations like this, when people’s lives are at stake. Statisticians and data scientists whose expertise lies outside epidemiology and public health problems (me included) will want to consider how the many dimensions of the COVID-19 pandemic intersect with our work and interests.

This is so much more than a biological problem, calling for research in many areas. Economic analysis is needed to address the important ways the pandemic will affect individuals, communities, companies, and their interactions in the US and around the world. Survey design and analysis will play an important role, especially for local areas and specific groups of people not included in large studies and general populations. Sociologists will consider the impacts on people’s lives and relationships. Environmental activists, education researchers, policy analysts, and others will study areas often overlooked in an immediate crisis that affect us in the long-term. Human rights mathematicians will look at how marginalized groups are being much more severely affected and the ways in which the crisis will be used to erode freedoms. (Just as one example, the early results of an analysis of county-level data by Peace-Work volunteers indicate a strong correlation between local COVID-19 fatality rates and poverty rates.)

There are ways in which everyone can be involved. All of us can point friends, businesses, organizations, and governments to accurate information to foster action based on data and sound analysis instead of fear, superstition, or propaganda. Never has our unique expertise been more valuable, and we can help set a course for local communities and organizations to rely on the best science and research in the future.

The COVID-19 pandemic will change our world and all our individual lives in ways we have not yet understood. Yet, one difference is apparent already. The D4G community—volunteers, professionals, academics, and students—is being seen by many people, often unconnected to data and analytics. The world is seeing us as a powerful force for the greater good—every day, but especially in time of crisis. This work is saving lives, and future generations will be inspired by what we are doing now. This will be our finest hour.

Beyond the present crisis, this moment in history is changing the future of our work. Critical infrastructure is being created and best practices established and shared. And all of us are more interconnected in our shared pursuit to use science to help people and save lives. Together, we are building the future of analytics—a new day in which our shared commitment to service for the greater good will touch everyone, strengthen lives and communities, and make our world a better place. This tragic pandemic will be a schoolroom for us, developing and refining the practices that will serve us well the next time!

Get Prepared to Get Involved
Normally, this column includes opportunities for getting involved—groups looking for volunteers, programs accepting applications, and so on. In this time of social distancing, now is a great time to focus on our collaboration skills. New resources are being added every day to support analytics across the miles. Use this time to make a new contact, attend a webinar, learn best practices for virtual communication. Reach out to that organization far away and the people you were hoping to meet someday. Check out their website, attend a webinar, send an email, and get involved in their important work. Always remember that collaboration is the life blood of scientific research. Use this time to make virtual connections with researchers outside your usual circle and make a difference for good.