The story continues - after @ikashnitsky and I pointed out that this paper was mathematically impossible, and had numerous errors, it was partially corrected
Now, the lead author is calling us "trolls" https://twitter.com/anya1anya/status/1355865046392430595
Now, the lead author is calling us "trolls" https://twitter.com/anya1anya/status/1355865046392430595
His initial response to us, when we privately emailed him about the issue in the paper, was, to quote, "you are not just right because you THINK you are" (caps=italics)
Now, remember, this is not a minor paper
Altmetric of 3,400, in dozens of news articles, and it's been cited by the WHO and EU already
This paper is impacting school reopening policy across the globe
Altmetric of 3,400, in dozens of news articles, and it's been cited by the WHO and EU already
This paper is impacting school reopening policy across the globe
Now, it's important to note that, as far as the @JAMANetworkOpen editors and the author was concerned, that was that. We were told to submit a 600-word comment online and if the authors wanted to, they could respond
After we persisted, publishing a lengthy critique of the paper https://osf.io/9yqxw/ and getting a bit of press, we were invited again to submit a 600-word comment and told the authors would have to respond
And so we did. Referencing our longer critique, we fit a few of the most egregious errors into 600 words and clicked "submit" on the 7th of December
Then we waited. And waited
Then we waited. And waited
It was a bit surprising for us that we had to wait at all. The @JAMANetworkOpen website proudly proclaims that comments will be put up after 2 days maximum, but a month after submission...still nothing
Finally, on the 8th of January, we checked back to see that both our comment and a response from the authors had been posted
Our comment was strictly limited to 600 words (inc. references)
The authors' response? Nearly 1,500
Our comment was strictly limited to 600 words (inc. references)
The authors' response? Nearly 1,500
And guess what? We were right
There was a clear mathematical error that completely reversed the results of the paper
So the authors rewrote the whole thing to try and salvage something from the ruins
There was a clear mathematical error that completely reversed the results of the paper
So the authors rewrote the whole thing to try and salvage something from the ruins
Now, instead of showing that school closures kill more people than COVID-19, the paper shows that school closures either kill nobody at all or countless millions depending on whether your kids are in the US or EU
I kid you not
I kid you not
For some context, the new figure of YLL that is caused by school closures if you cherry-pick just the results the authors would like to use means that 3 months of closed schools has cost more YLLs than influenza in the U.S.
Since ~2015
Since ~2015
I'm not going to go into just how awful the study is again, except to note that if you want your mind to boggle have a look at the threads below https://twitter.com/GidMK/status/1348441370180997123?s=20
But the real story here, to my mind, is that when junior colleagues CORRECTLY pointed out errors in a paper, they were ignored and belittled by editors and authors alike
For months
For months
Meanwhile, the paper has been cited a dozen times academically, is in planning and policy documents across the world, despite the fact that all of this happened when the work was quite simply wrong
Now, it's very hard to guess at the impact of such a paper, positive or negative
But it is entirely possible that this study has literally killed people through the errors of the authors
Who have called us "trolls" for pointing out the mistakes
But it is entirely possible that this study has literally killed people through the errors of the authors
Who have called us "trolls" for pointing out the mistakes
I should also note that Ilya and I get literally nothing for all of this. We have spent something like 50 hours each carefully identifying errors, writing letters, editing them down etc for no pay, mostly after work and on weekends