Tech & society

Misleading COVID-19 headlines, take 1,269 🤦‍♀️

News reports focus on daily COVID-19 data which does not necessarily a reflect transmission rates but the day of the week or backlogs. This example from Texas bears out the challenge.

Texas sees 1,000 new coronavirus cases 5 days in a rowThe Hill, 13 May 2020.

Texas began reporting more than 1,000 new cases daily on 30 April.

Only two days in the subsequent 15-day period have dipped below 1,000; six of 15 days, the number has been more than 1,100. Today’s numbers: 1,355 new cases.

The story isn’t Texas data for the past five days.

The story is that steadily increasing seven-day average you can see above, which started on 24 April.

Twenty days ago, on 24 April, the seven-day average for new cases in Texas was 766.

Today it was 1,140. The day’s total was 1,355.

Why use a seven-day rolling average instead of daily numbers?

To try to even out the artifacts related to testing reports that have nothing to do with actual new cases, such the number of people running testing centers and laboratories. Moreover, there is paperwork lag (submittal to lab test to report to state to report to state public database).

When is a focus daily numbers appropriate?

When there is a major aberration, that’s news and it deserves an explanation. Usually large changes result from adjustments in how data are being reported. An example is when New York added to death totals, including probable COVID-19 related deaths and not strictly confirmed-by-test deaths.

Originally published in my newsletter:

COVID-19 day 114 : 📈 1,390,746 cases; 84,136 deaths : 13 May 2020

📝 Subscribe to Kathy’s COVID-19 Memo 

By Kathy E. Gill

Digital evangelist, speaker, writer, educator. Transplanted Southerner; teach newbies to ride motorcycles! @kegill

Leave a Reply

This site uses Akismet to reduce spam. Learn how your comment data is processed.