Wednesday, April 15, 2020

Day 26: Corona: Lies, Damn lies and Statistics … and why the quip is also slightly off.

This article is part of The 100 Days Offensive. Got to Day 24 or Day 26a.

Mark Twain once said: “There are three kinds of lies: lies, damned lies, and statistics.” This seems currently to be truer than ever. It is very hard to find even two Corona statistics that do not seem to contradict each other.

We have several problems at the same time:
  • there is no unified model to even count the number of cases or deaths,
  • the demographics behind the statistics vary wildly,
  • we are just in the process of building a solid model,
  • people are cherrypicking among the available statistics according to personal reasons and
  • and a press trained over decades to go for click rates is unable to cope with the complexity.
There is no simple “truth” or correct statistic out there. Nobody knows exactly what we’re dealing with. All I can tell you that the baseline numbers are enough to worry me (and I am not much of a worrier).

We are still in the trial-and-error-phase concerning the measurement and the mathematical models. But with what we know about the pandemic, I think it is better to err on the side of caution.

If you want to understand why this is so hard, there is a good comic explaining it:

So when going back to Mark Twain, we find that the quip (in different words) goes even further back in time. That phrase can be found in different form within the science journal, Nature, in November 1885: "A well-known lawyer, now a judge, once grouped witnesses into three classes: simple liars, damned liars, and experts. He did not mean that the expert uttered things which he knew to be untrue, but that by the emphasis which he laid on certain statements, and by what has been defined as a highly cultivated faculty of evasion, the effect was actually worse than if he had" (Source). So as we go backward, the statement becomes more differentiated. It is deeply ironical that a statement from the late 19th century fits so well to our current conundrum.

P.S. As most people can count, you are probably aware of the gap between 24 and 26. I will try to make it up with a two post on a single day soon.

No comments:

Post a Comment