It won’t be approved for kids <6 months. I’ll turbo vaccinate once they hit 6 months. I’d prioritize covid vaccine over the standard 6 month vaccines tbh
1 Like
You were shown another case multiple times, I’m still waiting for you to give us your excuse for Church.
I looked at both versions and as far as I can tell, very little changed other than including six weeks more data in the published Science version. The figures are the same (although with updated data in the later version). The materials and methods sections look to be substantially the same, although the version in Science goes into more detail. The abstract in the Science version is more detailed. Science articles don’t break down into sections like Results or Discussion, so it reads a bit different. But looking at the article I think that the preprint was just straight up accepted. Not even accepted with minor revisions, just accepted. Then the authors spent the next few weeks rewriting the paper to fit with Science’s formatting and and style guidelines, and updating the figures with the latest possible data.
So if someone can point out the serious flaw in the preprint that was corrected in the published version three weeks later, I’d appreciate it. Here’s the two versions for reference.
Since I weighed in on example 1 and people are using these examples for why church should be moderated more strictly, I decided to go through example 2 as well. (Before yesterday I had zero knowledge of either incident, so this is all first impressions).
I do agree example 2 looks bad for church. Looks like he posted a link to an old pre-print with dubious data and in the comments to the study someone noted the authors had released a more recent study that contradicted (or maybe updated?) the results from the study church shared. Not sure I fully understand the back and forth about the comment/links - as church seems to have acknowledged the comment, but then there was some weirdness (possibly trolling) about the link to the new report. It looks like church found an old pre-print that supported some position he wanted to push, and shared it without any analysis of whether it was valid (which is bad posting). Had it been a new pre-print, I could see just posting it without analysis, but feel if you’re finding 6-month old pre-prints you should take a closer look before sharing. I guess it’s possible he knew about the comment/new report (which would be even worse posting), but that makes no sense to me as people would obviously find that comment.
If this is a pattern with church, then sure, he should get less leeway when posting bad studies. However, I still stand by my conclusion that for the first example, since church provided the link to the actual study (and within a few posts shared that it was from a Dr. John youtube), he didn’t do anything wrong in terms of his posting (getting information from Dr. John youtubes is an entirely different subject that I have no ability to judge as I have no clue who he is, but also one that I don’t see as a moderation issue). Also, since I’ve been mod, I’ve not had any reports of church sharing bad studies - but have gotten a number of reports when he shares BBC articles, which seem entirely fine.
Sorry - this is probably getting close to a tldr now - but I guess my conclusion is that if people want church to be held to a higher standard than other posters, I would need more support. While he’s clearly a bad poster in some aspects (including that he likes to troll), I also see people immediately jumping on even benign posts from him. For instance, in example one in I would have banned anti-vax dog (for the reasons it was ultimately banned) but also given a decent time-out to the user behind that account.
3 Likes
There is one difference in the figures, they added co-morbidity factors to the Science version in Figure 3. There is no figure 3 in the preprint. Two explanations for this, it was a required revision that was turned around insanely quickly to put out this very important paper, or they submitted a different version to Science than the preprint. I don’t know which is more likely.
edit: I think I misread this initially, Figure 2 is risk of catching covid and I think figure 3 is chance of death having caught covid. So it’s a whole new figure! I thought it was a modification of figure 2 just adding comorbidities, but it’s not. And it makes sense to add comorbidities to the having caught covid figure and not to the chance of catching covid figure. And this makes me think that they submitted a different version to Science and not the preprint, because figure 3 isn’t a revision they could just turn around in a week or two I don’t think. That’d be totally nuts.
This is false. Criticizing sources is absolutely reasonable, because no one has infinite time to analyze everything. There is an immense body of covid preprints out there, the vast majority of which went undiscussed here for good reason: being unimportant, being low quality, being fundamentally flawed, being to esoteric, etc. Knowing that a given preprint is being pushed by an ivermectin promoter and is getting amplified in anti-vaxx communities is absolutely worth knowing when deciding to take a given study seriously or not. You don’t get to ex post facto say the authors published a substantially improved paper in Science, ergo this one was also good. As a preprint, it was worth dismissing as chaff at least until obvious errors were corrected. It’s also worth wondering why Dr. John, anti-vaxxers, and churchill all elected to amplify this preprint despite its obvious flaw that had been publicly pointed out in a manner that they could have seen for themselves.
Along those lines, though, your notion that mere mortals cannot possibly criticize a paper published in Science, that they have to be trained in the nuances of the field and not just in a general biological sciences background, is staggeringly ignorant. At my graduate school, finding flaws papers in the one word journals (Cell, Nature, and Science) was literal course work for first year grad students. Yes, papers that had passed peer review. Everyone got at least one a week to sort through and spot flaws in, culminating in an oral exam on a paper. If you couldn’t spot flaws in Science papers as a first year grad student, you were kicked out of grad school entirely. You were unfit to join a lab or get to sciencing at all.
Criticizing papers is easy. It’s one of the easiest parts of science. Your statement about the bar being lower for covid is the understatement of the century. The easiest way to find a bad paper is to look for papers in the one word journals on hot topics. For my time doing this, those were anthrax (remember that scare back then?) and RNA interference (as people grasp for the coat tails of a Nobel Prize). Bad science can and does get published on the regular, even in prestige journals, and spotting it is often so easy a first year grad student can do it.
4 Likes
This is complete nonsense. Having been a reviewer for a Science paper once, I can tell you that only the most competent, highly-trained, and physically attractive experts are capable of serving in this capacity.
10 Likes
Out of genuine curiosity, can you direct me to more mentions of this obvious flaw anywhere else than a single tweet by an Israeli dermatologist (who has a patreon page and a shop where you can buy a logo t-shirt of him, I’ll link if anyone wants)? I did message Dr. Avi to see if he agrees with CN analysis of the paper, that was based on his one tweet that had 0 reply (well it had one reply he wrote himself, but i’ll keep that one for later). I’ll update if he replies.
It’s def true that some bad science get published. It’s possible, but EXTREMELY unlikely that a really bad paper was uploaded on Oct 14, completely re-written based on alleged public opinion, only then sent to Science editor (as you suggest happened), read by him, sent to two reviewers, returned with questions, changes and remarks answered by the authors, corrected, re-sent, read-proof and published within 3 weeks.
Here’s the far more likely scenario to what had occurred on that dark night in October:
- Churchill posted a paper
- You guys really wanted to discredit this paper as soon as possible
- ikes searched twitter for something and landed on that fashionable dermatologist who posted a one liner about it that he ended with his own reply - “Dr John is a clown”.
- everyone else parroted the “the paper is flawed” mantra, which I somehow seen here over and over as if “flaw” is some universally accepted measurement of science.
- The paper was never anywhere near “flawed”. It was probably a very decent piece of data collecting that was surprisingly quite accurate.
I don’t question that bad science get published. I question how this paper would be looked at if it wasn’t Church who posted it. I remain very unconvinced that this paper had some extreme flaw that was so easy to catch and embarrassing for anyone involved but yet so extremely easy to fix within an almost impossible fast turnaround.
If I was a mean person, I’d even consider that seeing so many people parrot the “the preprint was flawed” mantra is a form of misinformation.
3 Likes
Pretty sure everyone in this thread has answered this question for you.
And again, every figure that was in the preprint is in the Science paper, unchanged as far as I can tell except for updated data that goes six weeks in the future. The methodology looks to me to be the same. What was the significant flaw in the preprint that was fixed in the science paper?
It has something to do with a controlled group, according to Wookie. He never showed any proof that the flaw was fixed or that the version is vastly different as he claimed it was. I’m not saying it isn’t, but he merely said that as a fact without showing any proof from the two available prints.
1 Like
That’s what I was looking for by comparing the materials and methods section of the two versions. They seemed the same to me but I’m very much a total layman in this area so I easily could have missed something.
1 Like
Who says it needed a complete rewrite?
Good point, we’re just trying to figure out what the significant flaw was in the preprint was that was fixed in the Science version. So completely rewritten? Not necessarily. But if a big flaw in the preprint had to be fixed then it’s probably more correct to say there was a substantial revision, not necessarily a total rewrite.
CN did label it ‘useless’ which would lead most to believe large sections, at least, require substantiial amendment if any of it was indeed still usable.
1 Like
Damn, it sounds like CN is being hyperbolic on the internet! Guards, seize him!
heh, reminds me of the time Jal posted that UK experts stating AZ gave (as good as) no protection after x months and the thread guards really did go ballastic! Jal said this!
Hyperbole… noted.
Last thing I’d want is a doc that practiced hyperbole.
1 Like
As far as I can tell, CN has done back-of-the-napkin work due to the supposed “flaw” and reached the conclusion that Pfizer efficiency was 70-82%, in line with other published work, once accounting for the asinine flaw.
The published work on Science have Pfizer efficiency at 45% if I understand correctly.
While I might be wrong in my understanding of his post and he did say it was rough work, it appears the hyperbole continued to his numbers that were not challenged.
1 Like
Whoa, a wookie responded? The plot thickens…
1 Like