It happens to all of us measurement types, sooner or later. You’ve got a project up and running and the data is coming in and things seem to be going fine. But then you review the purpose of your project and compare it to the numbers so far and then it hits you: “Uh oh!” Despite our best planning, sometimes we don’t really understand the flaws in a project until we look at the data.
Well, the the big league version of that just happened to the CDC and their COVID-19 data. The Atlantic reported this week that “The Centers for Disease Control and Prevention is conflating the results of two different types of coronavirus tests, distorting several important metrics and providing the country with an inaccurate picture of the state of the pandemic.” By combining figures for diagnostic tests and antibody tests, the CDC has been making it look as if they were doing more testing for active infections than they actually were. The CDC “…is mixing the results of viral and antibody tests, even though the two tests reveal different information and are used for different reasons.”
Says Katie Paine, publisher of The Measurement Advisor, “It’s a classic example of what you don’t do with measurement. The disease equivalent of combining impressions with likes and opens and views or visits.”
This is how trust in a brand is lost. The CDC is who we in the U.S. have always trusted to keep us safe. Alas, no longer. Individual states have used the CDC’s bad data to set guidelines for reopening. So here is yet another example of how an incompetent government puts its citizens in danger (see “White House Economic Advisor Kevin Hassett (and Anyone Else Who Uses Data to Kill Us) Are the Measurement Menaces of the Month”).
To be fair to all of us who make the occasional measurement mistake, the CDC is in a whole higher level of institutional incompetency. There were plenty of irregularities, political pressure, and mistakes before this most recent revelation. Amid all the incompetence and lack of leadership in the U.S. response to the pandemic, the CDC’s dirty data is just one more clown car in a parade of ineptitude.
If you do screw up like the CDC did, the very witty folks at Crooked Media have provided a wonderful example of an apology letter: ∞