March 22, 2020

Every night, usually around the time I go to bed, I get a recap of the day’s COVID-19 news from Public Health — Seattle & King County. The two key statistics in this message, the number of confirmed cases and the number of deaths, jump out at me every day. As of last night there are 934 total cases and 74 deaths, up 141 and 8 respectively from the prior night.

Tableau, my employer, is making COVID-19 data available for free. Many other companies and institutions are doing the same. My employer also enlisted Amanda Makulec, Operations Director of the Data Visualization Society, to publish on Tableau’s official blog an article entitled “10 considerations before you create another chart about COVID-19.” In this and other articles, Ms. Makulec cautions data scientists against creating visualizations that make misleading or unsubstantiated claims. As surprising as it is that a data visualization company would urge restraint in using its products, I’m glad that Ms. Makulec wrote what she did, and I hope that others listen.

During this time of crisis I’ve loosened my personal restrictions on social media usage. Normally I’d only check services once a day, but now there is simply too much going on at too many levels of specificity. To give an example of how specific social media have become, someone living near me set up a Facebook group for a part of Ballard spanning only three avenues and three streets. It in turn links to a shared Google Sheet with neighbors’ names, contact information, special needs, and offers to help. Discussions have included the closure of a local playground and tips on the least crowded time to shop at our various local supermarkets. We’re also talking about setting up a Zoom chat to see each other face to face — an odd yet comforting idea for people living within shouting distance of each other.

People with backgrounds in computer science, mathematics, statistics, and even marketing and growth hacking are now all of a sudden interested in epidemiology. I’ve seen plenty of models and charts created by people applying textbook statistical models to existing data. Many of these are unhelpful — one simple model shared to my university’s alumni group claimed that 440 million people in the United States would be infected by mid-July, more than 100% of the population — and wide distribution neither conveys authority nor proves correctness. Being curious about data and visualizing data are useful skills, but as general-purpose social media have shown over and over again, it is often counterproductive to share everything that’s on your mind at the moment.

Similarly, the language of business does a poor job of describing a response to an outbreak. LinkedIn, which I’ve historically respected more than most social media because of its honest and singular purpose (to get you your next job), has featured such stories on my news feed as:

  • Inspiring-looking photo essays about brave people in China going to work during the spread of COVID-19.
  • An English-language article by a Chinese-based Italian-born communications specialist complaining about media bias against China, including many factual errors about this and prior epidemics, whose author engages in angry ad hominem attacks in the comments against people (including a co-worker and myself) who point out these errors. That article had been viewed about 2 million times and shared many thousands of times the last time I had seen it.
  • Copy-and-pasted lists of inspirational stories about products and companies that thrived during periods of economic uncertainty and downturn. Drawing inspiration from only the success stories is a classic example of survivorship bias.
  • Fist-pumping celebrations of Amazon’s resiliency by David Glick, an 18-year Amazon veteran who is now Chief Technology Officer of FLEXE, a Seattle-based logistics firm. Glick’s breathless boasting about Amazon shows just the sort of non-empathy and win-at-all-costs culture that has made Amazon both fiscally successful and culturally challenged. It definitely makes me think twice about FLEXE, a company that often sends recruiter mail to me, as a potential employer.

Everyone responds to a crisis in different ways. My goal is to rely on fact-based and reliable sources. That means I’m not sharing or even reading those 800-word essays by a friend’s doctor’s sister’s cousin’s Italian study-abroad roommate. That means I’m not sharing or reading infographics from lifestyle magazines. That means I’m not embracing false hope or twisted bright-side takes, such as a “good news” list stating that Italians are only dying because they’re the oldest people in Europe (they’re not) or a travel blogger’s note that Anne Frank lived in an attic for 2 years, so you can survive your home quarantine. That also means that when I see misinformation being spread, however positive the intent, I’ll point it out. I urge others to do the same. This is a time for people to come together for good.