The problem with this isn& #39;t that it& #39;s mathematically bad, it& #39;s that it& #39;s appropriating mathematical ideas (wrongly) to suggest some certainty and science behind it. There& #39;s an entire cottage industry of this in newspapers (see Ben Goldacre passim) 1/4
If the aim was illustration, this does it better. It says that two factors contribute, but is ambiguous how. That& #39;s fine, of course. More than this should contribute to that level, and it can be dynamic. But the basic idea is this. 2/4
More concretely, something like this gets the idea across. There are no numbers on it, but that& #39;s an issue of detail and policy -- and we don& #39;t know what that is. We use graphs like this all the time when weighing up risks. 3/4
This is concerning because no-one scientifically literate would jump to the first one. That& #39;s several layers of dumbing down while trying to look smart. 4/4
You can follow @DrCDArmstrong.
Tip: mention @twtextapp on a Twitter thread with the keyword “unroll” to get a link to it.

Latest Threads Unrolled: