Data in the Time of COVID

In beginning of the COVID crisis that evolved from news reports out of China to a PHEIC to a pandemic, data has been evolving as rapidly. Initially whate were patchy, spotty, slow-moving streams improved to an ever-increasing, better managed, timely torrent. These page complained of the data at the beginning, but over time, vast improvements have been made.

All over the globe, we see the benefit of pbulic goods, such as the branches of the CDC, the WHO, government health and public health agencies and the universities, all gathering and disseminating valuable data, in as close to real-time as they can.

I have no inside look at the data collection and management systems, but from my view on the outside, looking for publically available data, Canada and Ontario and the other provinces have all stepped up their game, both in quality and timeliness. But, the universites have really kicked in. All the public money spent in universities has been returned manifold when one looks at the UofT Dalla Lama school covid website, or Johns Hopkins University tracker or the data tools coming out of Oxford. It is worth noting that a lot of this builds on open-source tools such as free database software or the core technology of the software industry, git. Much of the visualization tools probably were developed to manipulate (big) data for surveillance capitalism system, but a lot of it is free to get and use.

Thus, a computer literate working at home, can putter away, downloading datasets that count cases, deaths, vaccinations, recoveries and timestamp the records. And, one can see where the TV and radio news reports are getting their info, and whether there is biases in the reporting.

Of course, the data is only as good as the original information coming from the field, and we can see that it is mightily flawed at times. Witness the UK dropping 50K cases due to an Excel issue, or Russia finally admitting they undercounted deaths by a 100k. It will be interesting in the future for the data wonks to analyze how good the data was for any particular jurisdiction on some time period for some variable. That is the stuff of master's theses of the future. But, if one assumes the data is a very good guide to the trends of the virus, though not perhaps the absolute numbers of cases or deaths, we have been well-served by the public funded institutions.

GE - some notes