4 ms·

We've reached the point where science is totally broken and now entirely partisan. I've read his papers. I've also read papers by other epidemiologists. There's

We've reached the point where science is totally broken and now entirely partisan. I've read his papers. I've also read papers by other epidemiologists. There's no comparison: his are far better. This is not necessarily a compliment, because the field of epidemiology is atrocious. You are presumably flaming him over his IFR estimates. Those estimates come from a meta-study of serosurveys, thus can at least theoretically measure mild cases where the infected person didn't get tested. And it's a meta-study, so hopefully captures plenty of real world data. Do you know what Professor Ferguson's team at Imperial College London use as their own IFR estimate? They're still citing their own Verity et al paper, whose IFR estimate of 1% only used data from January/February 2020 and which was based on media reports coming out of China, along with a few flights of Wuhan evacuees. That's it. That's all it's based on. Despite this, many other epidemiologists are also still citing the ICL numbers despite them being long since obsolete and using a tiny sample size. If you're so convinced Ioannidis is such a terrible scientist, please explain: 1. Why his approach is unacceptable but ICL's is OK? 2. Why was his reputation pre-COVID just fine? He's actually quite famous for pointing out low standards in other scientific work, isn't he?

epistasis 5y ago
Science is just fine. This guy is not. Part of keeping science fine means eliminating those who prioritize politics over the search for the truth. I've only reviewed Ionnaddis's paper, with a quick literature search, found it to be hilariously fraudulent. I haven't investigated ICL's analysis, but if you can be more specific about what to look for, I might. For 2., his work was about the supposed "low standards" of others, but IMHO it was always coming from the place that every paper in the literature has to represent pure truth, and every study should have huge numbers before getting published. Though I disagree with that sentiment (I think the literature should publish early and often, like open source), it's a fair enough one to have, which is what garnered him fame and respect. To see his own research have such openly low standards, when trying to contradict consensus, shows an arrogance and lack of self-skepticism of the sort that means I would never be able to trust him as a collaborator, and that I won't trust his judgement and even honesty in the future.
- nradov 5y ago
  Scientific fraud is a serious accusation. Which specific part of this peer-reviewed study is fraudulent? Have you contacted the WHO and asked them to retract it? https://pubmed.ncbi.nlm.nih.gov/33716331/ https://pubmed.ncbi.nlm.nih.gov/33716331/ https://doi.org/10.2471/blt.20.265892 https://doi.org/10.2471/blt.20.265892
- native_samples 5y ago
  Sure. Here's one example article reviewing a recent paper from ICL: https://dailysceptic.org/2021/08/24/examining-the-latest-paper-from-imperial-college-london/ https://dailysceptic.org/2021/08/24/examining-the-latest-pap... The same site has a lot of paper reviews, especially focusing on that team (because they are influential). Look at the sidebar section under "How reliable is the modelling". What you may be doing here is reading Ioannidis' papers and saying, wow, these are total crap. Therefore, he is a bad scientist. This would be an easy mistake to make if you aren't familiar with the low quality of the scientific literature. I'm here to tell you that his papers are not that great, but his competition is far worse. It's not that JPI is a bad scientist, it's that "good scientists" that meet the basic thresholds we've all been socially conditioned to accept hardly seem to exist in the wild. This does not necessarily make Ioannidis a hypocrite or fraud. It kind of does, but the problems he identified are all structural and to do with bad academic incentives. Yet, Ioannidis is himself an academic. He is working within the system that creates those bad incentives, and he is not immune to them either. It probably isn't possible to do great science in academia in fields like epidemiology, where experiments aren't possible, but at least he gives the impression he's trying. Other teams are full blown political activists pretending to be scientists, like the paper I linked to above, where the first sentence is a lie about publicly available statistics and the entire goal of the paper is to tell policymakers what to do.

epistasis 5y ago

Science is just fine. This guy is not. Part of keeping science fine means eliminating those who prioritize politics over the search for the truth. I've only reviewed Ionnaddis's paper, with a quick literature search, found it to be hilariously fraudulent. I haven't investigated ICL's analysis, but if you can be more specific about what to look for, I might. For 2., his work was about the supposed "low standards" of others, but IMHO it was always coming from the place that every paper in the literature has to represent pure truth, and every study should have huge numbers before getting published. Though I disagree with that sentiment (I think the literature should publish early and often, like open source), it's a fair enough one to have, which is what garnered him fame and respect. To see his own research have such openly low standards, when trying to contradict consensus, shows an arrogance and lack of self-skepticism of the sort that means I would never be able to trust him as a collaborator, and that I won't trust his judgement and even honesty in the future.

nradov 5y ago
Scientific fraud is a serious accusation. Which specific part of this peer-reviewed study is fraudulent? Have you contacted the WHO and asked them to retract it? https://pubmed.ncbi.nlm.nih.gov/33716331/ https://pubmed.ncbi.nlm.nih.gov/33716331/ https://doi.org/10.2471/blt.20.265892 https://doi.org/10.2471/blt.20.265892
native_samples 5y ago
Sure. Here's one example article reviewing a recent paper from ICL: https://dailysceptic.org/2021/08/24/examining-the-latest-paper-from-imperial-college-london/ https://dailysceptic.org/2021/08/24/examining-the-latest-pap... The same site has a lot of paper reviews, especially focusing on that team (because they are influential). Look at the sidebar section under "How reliable is the modelling". What you may be doing here is reading Ioannidis' papers and saying, wow, these are total crap. Therefore, he is a bad scientist. This would be an easy mistake to make if you aren't familiar with the low quality of the scientific literature. I'm here to tell you that his papers are not that great, but his competition is far worse. It's not that JPI is a bad scientist, it's that "good scientists" that meet the basic thresholds we've all been socially conditioned to accept hardly seem to exist in the wild. This does not necessarily make Ioannidis a hypocrite or fraud. It kind of does, but the problems he identified are all structural and to do with bad academic incentives. Yet, Ioannidis is himself an academic. He is working within the system that creates those bad incentives, and he is not immune to them either. It probably isn't possible to do great science in academia in fields like epidemiology, where experiments aren't possible, but at least he gives the impression he's trying. Other teams are full blown political activists pretending to be scientists, like the paper I linked to above, where the first sentence is a lie about publicly available statistics and the entire goal of the paper is to tell policymakers what to do.