"The plural of anecdote is not data."
The methodology of collecting anecdotes matters immensely. Merely collecting anecdotes based on whoever manages to comment in a random thread opens your collection process up to massive selection bias. In fact we even have a name for this process: a filter bubble. Every filter bubble in existence is the result of people assuming that not only is the population of their bubble representative of the average person (it's not), but also that people on either side of an issue will be equally likely to offer their anecdote (they're not).
Without rigorous collection methodology, anecdotes do not sum to data.
(Note that this comment says nothing about the broader topic of omicron severity; I don't have data, listen to people who do.)
But that's not as pithy.
> you can collect 10000 anecdotes, put them in a spreadsheet with some information about the people who said it, and suddenly you got data.
You can, but that's not a controlled study. The anecdotes can't be verified. So that data is not worth citing. I wish it were, which would make science a lot easier.
To add another link to the one Bumby has http://blog.danwin.com/don-t-forget-the-plural-of-anecdote-i...
which actually it makes a lot more sense to me this quote than the misquote because >You can’t collect a couple of anecdote and pretend it’s data.
sure but you can collect 10000 anecdotes, put them in a spreadsheet with some information about the people who said it, and suddenly you got data.
I mean that is basically what I thought to myself every time somebody said the plural of anecdote is not data but bit my tongue because not wanting to get into a war over it, and now I find out the original was actually exactly what I thought it should be.
Thanks munificent! Your name certainly applies for me.