2 ms·

> What is your basis for this claim? Well you answered your own question: It is true that the biggest funders (NSF, NIH) are not market-focused. A lot of resea

> What is your basis for this claim? Well you answered your own question: It is true that the biggest funders (NSF, NIH) are not market-focused. A lot of research is funded by taxes. That's an exclusion from market mechanisms. They don't have to convince the actual consumers of the research to buy it, we are all collectively forced to buy it by law. > No field I know of out-right tolerates fraud I know quite a few such fields, so we might have a different definition of "tolerate". After all this story contains the following paragraph: “It’s unfortunate that it has taken 2 years to make the decision to retract,” says Donna Wilcock, an Indiana University neuroscientist and editor of the journal Alzheimer’s & Dementia. “The evidence of manipulation was overwhelming.” We're talking about a retraction here, which is the weakest response possible. So ... it took two years of "investigation" to do nearly nothing, after other people did all the investigative work for free, and one of the authors continues to be employed with no consequences whatsoever even though his co-author admitted the figures were tampered with. I'd argue this is what institutional tolerance of fraud looks like.

DiogenesKynikos 2y ago
There are no "consumers" who "buy" basic research. You're talking about fundamental research into basic scientific questions as if it were the same as potato-chip manufacturing.
- mike_hearn 2y ago
  The "consumer" who "buys" research in a commercial setting is usually the executive who is funding the department. In that context stuff like P-hacking, HARKing etc doesn't happen much because at some point your bosses boss is going to read your internal paper and notice that your claimed discovery has nothing to do with what you were originally asked to investigate. In academia that doesn't matter, it still counts as a discovery because nobody is really checking your pre-registrations. In corporate research it'll either be checked by the people paying your salary, or at a larger scale, it'll be checked by a regulator who is forcing you to pre-register your clinical trials.
  - DiogenesKynikos 2y ago
    There's no "consumer" who "buys" research into whether protons decay. That's basic research that will only occur if funded by national agencies that are not motivated by return on investment. As for commercial research, similar problems of fraud exist as in academic research. Instead of prestige, the motivations are things like bonuses and promotions. Academics actually care a great deal about fraud, funding agencies hate fraud and punish it, journals hate fraud - everyone dislikes it. Competing labs have every incentive to catch fraud conducted by their rivals. The idea that academia is rife with fraud and that nobody cares is just not true. There will always be a certain level of fraud not just in research, but in every economic sector, every intellectual pursuit, and every sport, commercial or not. There's no system that will perfectly eliminate it, but because of its empiricism and openness, science is fairly good at correcting itself over time.
    - mike_hearn 2y ago
      Physics gets large amounts of commercial investment. Quantum computers have been invested in heavily by the private sector for years. The defense world invests in laser research, etc. Plenty of examples out there. You can always find a topic you personally feel is super important even though it has no practical applications today, but such arguments are unfalsifiable. I can just as easily argue [more] proton decay research doesn't matter [yet] and you can't prove I'm wrong, so it's not worth going there. The same problems exist but not on the same scale. If your employer discovers you committed fraud to get a promotion you will certainly be fired and quite probably be taken to court by their legal department. If the fraud is at the level of the company they risk destruction and imprisonment by the government. It's not like in academia where they'll sit on it for years and then, maybe, request a little notice to be put on the paper's web page - all without the government even noticing let alone caring. The huge difference in consequences yields different risk/reward ratios and that's reflected in how often these problems are found. > Academics actually care a great deal about fraud, funding agencies hate fraud and punish it, journals hate fraud - everyone dislikes it. Do they? How is the co-author of this paper still employed if funding agencies punish it? Where are the university funded research-police departments? Why do we keep hearing cases like this Alzheimer's one and why are there never any announcements by Vice Chancellors about doubling investment into fraud investigations as a consequence? Why is a research audit not something that these fraudsters fear? How is it the case that publishers discover after the fact that dozens of their journals have been completely compromised by paper mills, instead of it being discovered via some more active process before publication occurs? > science is fairly good at correcting itself over time Absolutely not. If science was good at self correcting it wouldn't take nearly two decades for someone to notice that a widely cited paper was forged, and the people who notice these things wouldn't need to be anonymous. But they do. Look at the Gino case. She launched a massively well funded lawsuit against the people claiming she engaged in fraud. That's the exact opposite of self correction.

DiogenesKynikos 2y ago

There are no "consumers" who "buy" basic research. You're talking about fundamental research into basic scientific questions as if it were the same as potato-chip manufacturing.

mike_hearn 2y ago
The "consumer" who "buys" research in a commercial setting is usually the executive who is funding the department. In that context stuff like P-hacking, HARKing etc doesn't happen much because at some point your bosses boss is going to read your internal paper and notice that your claimed discovery has nothing to do with what you were originally asked to investigate. In academia that doesn't matter, it still counts as a discovery because nobody is really checking your pre-registrations. In corporate research it'll either be checked by the people paying your salary, or at a larger scale, it'll be checked by a regulator who is forcing you to pre-register your clinical trials.
- DiogenesKynikos 2y ago
  There's no "consumer" who "buys" research into whether protons decay. That's basic research that will only occur if funded by national agencies that are not motivated by return on investment. As for commercial research, similar problems of fraud exist as in academic research. Instead of prestige, the motivations are things like bonuses and promotions. Academics actually care a great deal about fraud, funding agencies hate fraud and punish it, journals hate fraud - everyone dislikes it. Competing labs have every incentive to catch fraud conducted by their rivals. The idea that academia is rife with fraud and that nobody cares is just not true. There will always be a certain level of fraud not just in research, but in every economic sector, every intellectual pursuit, and every sport, commercial or not. There's no system that will perfectly eliminate it, but because of its empiricism and openness, science is fairly good at correcting itself over time.
  - mike_hearn 2y ago
    Physics gets large amounts of commercial investment. Quantum computers have been invested in heavily by the private sector for years. The defense world invests in laser research, etc. Plenty of examples out there. You can always find a topic you personally feel is super important even though it has no practical applications today, but such arguments are unfalsifiable. I can just as easily argue [more] proton decay research doesn't matter [yet] and you can't prove I'm wrong, so it's not worth going there. The same problems exist but not on the same scale. If your employer discovers you committed fraud to get a promotion you will certainly be fired and quite probably be taken to court by their legal department. If the fraud is at the level of the company they risk destruction and imprisonment by the government. It's not like in academia where they'll sit on it for years and then, maybe, request a little notice to be put on the paper's web page - all without the government even noticing let alone caring. The huge difference in consequences yields different risk/reward ratios and that's reflected in how often these problems are found. > Academics actually care a great deal about fraud, funding agencies hate fraud and punish it, journals hate fraud - everyone dislikes it. Do they? How is the co-author of this paper still employed if funding agencies punish it? Where are the university funded research-police departments? Why do we keep hearing cases like this Alzheimer's one and why are there never any announcements by Vice Chancellors about doubling investment into fraud investigations as a consequence? Why is a research audit not something that these fraudsters fear? How is it the case that publishers discover after the fact that dozens of their journals have been completely compromised by paper mills, instead of it being discovered via some more active process before publication occurs? > science is fairly good at correcting itself over time Absolutely not. If science was good at self correcting it wouldn't take nearly two decades for someone to notice that a widely cited paper was forged, and the people who notice these things wouldn't need to be anonymous. But they do. Look at the Gino case. She launched a massively well funded lawsuit against the people claiming she engaged in fraud. That's the exact opposite of self correction.