According to a new science paper:>Heathers (2024), How Much Science is Fake? Approximately 1 in 7 Scientific Papers Are Fakehttps://osf.io/s4gcehttps://doi.org/10.17605/OSF.IO/5RF2MMedia story:>One in seven science papers is not to be trusted, says new science paperhttps://www.SmH.com.au/national/one-in-seven-science-papers-is-not-to-be-trusted-says-new-science-paper-20241018-p5kjfj.html
>>16442461The paper (links in OP)
7 in 7 are fake
So this is a science paper that encourages not trusting science papers, including the very same paper? So science is trustworthy. Carry on.
>>16442461
>>16442462
>>16442461They're all fake now. There's no paper involved at all anymore.
>>16442461Only?
>1/7bump it up to 5/7.1/7 is old news reformattedother 1/7 is actual science
>>16442552This. Just check the source and avoid the paper mills mentioned in this paper.>>16442512>>16442582>>16442600>>16442600>>16442602>>16442605This thread is a dog whistle for science deniers.
>>16442461Reproducibility crisis. Science is a constant method, not a one and done type of thing. Biology evolves because of quantum physics, dna, and white holes/black holes altering reality. That means sometimes experiments will not be able to reproduced and sometimes they were straight up lies for the scientists numbers to go up like manipulating a stock.
>>16443160Yeah black holes exist, that means white holes exist, white holes are mathematically proven to exist. But since black holes exist anyway you don't even need to prove white holes exist because of sacred geometry and as above so below geometry. Microcosm and macrocosm.
>multiple analyses suggest 15% or more of published science is fake or fraudulent>Option 1: Acknowledge problem and try to do better>Option 2: Deny, cope, seethe, and threaten the non-believers with an eternity in science hell
>>16443308Option 3 which is the easiest one. Let people publish whatever the fuck they want and stop with the science monopoly. Let brainlets die for being low IQ monkeys, let high IQ chads survive
and i bet 99.99% of the fake paper slop from china or india. not a hit for science, just another confirmation that 3rd-world garbage is polluting the western world
>>16442552only about 6/7 of the time
>>16443536If you read the paper the estimates come from studies only looking for the types of problems detectable by someone without intimate knowledge of the field or specific research, for example shooped or reused images or data that's mathematically impossible. The actual amount of fraud is almost certainly much higher.
So that paper can be trusted at a rate of 1 in 7. Nice.
>>16442512thats only 99.999999% tho
>>16443166>white holes are mathematically proven to existstring theorists come up with all sorts of equations but literally ZERO evidence of strings or any of that crap has been found in decades
>>16442461This is a meme or what?
>>16443337>fell for the survival of the fittest memekek. i mean yea its true but youll only wind up with retards in the end.
>>16442461can we trust op?
>>16443160We're not allowed to publish results from reproducing science if it refutes an Israeli scientist
>o fug this study casting doubt on other studies might embolden the chuds>not to worry I'll use the brilliant word games I learned in journalism school to deboonk it with its own conclusion!>but wait, if the study is fake then it must mean my deboonk is fake meaning the study is real but if the study is real then the deboonk is real meaning the study is fake but if the study is fake then...AAAAHHHH WHAT DO WE DO SHILLBROS??
But I can trust your paper though, right?
>>16442461It's FAR higher than that.
>>16442552Just like how paleontologists faked Archaeoraptor but Tetrapteryx- EXCUSE ME! Microraptor is still totally legit.
>>16442552It said 1 in 7 are NOT to be trusted. There is a 6/7 chance this is a trustworthy paper.
>>16442461>ONE IN SEVEN SCIENCE PAPERS FRAUDULENTIt's because 57.4 percent of all statistics are made up.
>>16442461Politically motivated/ideology poisoned science.
>>16442461Is this at all surprising?Virtually anyone I have known who is really really knowledgeable in their field will say most people in their field are full of shit. In and out of academia.
>>16442552the whole point is that you need to look for consensus and reproducibility. You need to see your claim evidenced by multiple lines of evidence across contexts, funding sources, etc. People who pull out a single study trying to claim something are pseuds.
>>16453363Atheists can't do science because atheists can't be honest. Atheism is a subset of communism, so its not ideology in general, its one particular ideology. Its not like this same pattern of events didn't all play out before in the USSR.
To be fair all of science is an endeavor to be less wrong over time. Consensus over even major stuff had flip flopped multiple times in the past. The only area I can think of that is yet to receive its "actually..." gotcha is thermodynamics.
>>16457702you don't understand. It's not a blind consensus. Every single scientific paper is an argumentative work. You need to evaluate their methods and the data presented and decide if you agree or are convinced. If something is consistently evaluated properly, replicable, and observed across contexts, it becomes irrefutable due to the body of vetted evidence (that scientists are trained to evaluate). We don't even give heed to the "discussion" sections of the author's interpretation of their results; we focus on the raw results themselves. The outcomes of a single paper mean nothing to us; it only raises suspicion for further research to be done on the subject. The example of global warming hysterics is in the minority and riddled with poor quality evidence compared to the vast towering data supporting otherwise.
One in seven people is Chinese
>>16443337What's stopping anyone to publish? Just host the articles on your own website, or web service
>>16442461What if I disregard Indian and Chinese papers? What is the ratio then?
bump
No. NO. NOOOoooooooooo TRUST THE BASEDENCE GOYS!!!
>>16445845by my calculations the irony is almost certainly lost on approximately 102% of /sci/, regardless of meme status
>>16442461> one in sevenit is affraidhttps://www.youtube.com/watch?v=gNtQmpcNge0#t=1m40s > at least half, if not 80 to 90 percent of the papers published in this field, are wrong.
>trust the science>the science tells you to not trust the science
>>16466703It always circles back to Chud Derangement Syndrome with the Grauniad.>th-the chuds are using the science wrong, s-stop!
>>16442461Artificial intelligence was a mistake, I bet more even more papers today are written either completely, or with some assistance of a large language model tool.
>>16474249TSMT1 in 7 is known to be wrong, the other 6 are only suspected of it
>>16442461 a complete universal field model. I submit this video for peer review. https://youtu.be/noBldW3A5IU?feature=shared
I just started reading this paper and I bet that the huge majority of falsified research is going to be of the statistical analysis type — because it’s easy for 3rd worlders to massage and fake data sets to get a desired result or just to outright plagiarize someone else’s results.
Super interesting little tidbit — there are some data forensics people who think the amount of faked work is MUCH higher than the median estimate. That’s interesting because in my experience people who confidently call outlier figures are the best, most dedicated to their field and don’t mind calling the high number where the field prefers to play it safe. Also going by the author’s methodology he’s probably under-shooting the actual rate of fakery — and I bet he thinks so too based on the fact he decided to include the highlighted snippet at all.
>>16478554Spoke too soon, the author concludes admitting that his methodology should be seen as a MINIMUM and the actual fraud level is much higher.Wow boys, the jeet hell is the real deal.
>>16442461One is seven people are Indians... coincidence?
>>16478563and another 1 in7 are Chinese
>paper millsmust be even worse at the big money owned journals like nature and science
>>16442461>Ctrl+F "Ioannidis">0 results
>>1644246185.7% of the time, it works every time.
>>16442461>Approximately 1 in 7 Scientific Papers Are FakeThat's not what the study says chud. Do you have the literacy of a goldfish?
this sentence is false
>>164843451/7 is within the confidence interval of the estimate, and is an interesting numberstop posting, boor
>>16478563Concentration is important
>>16456564>Muh consensusFound the brainlet. Caring about not upsetting the consensus is the reason all this fake science is published in the first place. Some inbred jew like Einstein comes along, shits out some mathematical diarhea on a platter, then the jewish press take it up and declare it the new revolutionary theory of physics, some faggot who wants a nobel prize makes an experiment with fudged numbers, the theory becomes consensus and if you publish anything refuting it you are cast out.After that of course, plenty of scientists try to prove the theory, but the data doesn't agree with the theory, so the study goes unpublished, it's not good to publish science that goes against the council of truth after all. Then some hack comes along with more tortured numbers, further reinforcing the religious belief. And then, of course, when the data is so outrageously against your theory that no fudging will do the job, all you have to do is invent a massless, interactionless particle and tell people to trust le heckin science! Or did you find out that your relativistic gravity fails at predicting the trajectories of stars? No problem, just invent "dark matter" and say that it comprises 90% of all matter, just that it can't be observed so don't forget to trust le science and saint Einstein! Huh, what did you say, particles don't show mass increase in particle accelerators? Well, that's easy, just say that the mass is relativistic, so there's no actual increase in mass, but it's there, somehow even if that makes no logical sense, trust us!
>>16493699>but the data doesn't agree with the theory, so the study goes unpublishedyou've clearly never been in academia. Results that don't go along with accepted theory are much more likely to be published in even higher-impact journals, because they want a groundbreaking cover story.>some hack comes along with more tortured numbers, further reinforcing the religious beliefPeople aren't failing to replicate results and then fudging them at mass to support a theory. If anything, when a result isn't replicable, it calls into question the validity of the initial result and starts a frenzy of people trying to out the misconduct. If something is generally well accepted, it's because there's a wealth of high-quality evidence to support it. If one bad actor fudges the numbers (which does happen), I can assure you that others will not blindly fudge their numbers to fit the theory. Instead, the irreproducibility of the finding raises mass suspicion and frustration from those trying to reproduce the results. It’s hard to see how these kinds of grand conspiracies you claim could function in practice.
>>16494767>because they want a groundbreaking cover story.So if someone debunked the theories that claimed that racial and sexual inequalities were the product of oppression, and not innate differences, they would face little to no obstacles?
>>16494779Yes, they would face a few obstacles because the consensus is already that it's an interplay of both nature and nurture. >debunked the theories that claimed that racial and sexual inequalities were the product of oppressionthere's nothing to debunk here. There's thousands of papers supporting the existence of both, and thousands demonstrating the interplay of sociologic and historic contexts. If you were to claim that it's SOLELY innate differences, you would receive flack from the scientific community because you clearly don't know how to compile and interpret evidence and have a cursory understanding of human biology.
>>16494770nearly a third of the papers in Nature or Science cannot be replicated because they prioritize extremely shocking results, which is fine because the whole reason they're disseminated like that is first to raise suspicion about something that could be groundbreaking and prompt further research. Then the community evaluates it further, realizes it was a dead end, and moves on. This is why quoting findings from a single paper is the peak sign of a complete pseud with zero academic experience or rigor. The purpose of a single study is to raise suspicion for further research, not act as groundbreaking evidence. Plus, academic papers themselves are ripped apart by peers even after the fact. It's a persuasive essay, not doctrine. Have you ever attended a journal club full of scientists? Nothing is taken at face value, and every claim the author makes and every method used becomes a point of contention. But no, you've never attended a journal club with scientists, you've never peer-reviewed an article, you've never responded to reviewer comments and had your work published, you have no clue about this process because you're not a scientist.
>>16494808>But no, you've never attended a journal club with scientists, you've never peer-reviewed an article, you've never responded to reviewer comments and had your work published, you have no clue about this process because you're not a scientist.Damn... you didn't have to do him like that, anon.
i did not know the onion was still around
>>16442461A man I respect way more tells me not to trust his paper.https://www.youtube.com/watch?v=gNtQmpcNge0#t=1m40s > at least half, if not 80 to 90 percent of the papers published in this field, are wrong.
>>16442461all those that fits your bias, of course.
>>16494767>you've clearly never been in academia. Results that don't go along with accepted theory are much more likely to be published in even higher-impact journals, because they want a groundbreaking cover story.This is a lie. If you go against certain "established" theories, your paper won't even get the option of being published. >People aren't failing to replicate results and then fudging them at mass to support a theory.Except when you go against one of these taboo theories, then nobody will publish a replicate that does not fit. Only those that fudge the numbers to get what the cult expects get published.
>>16495171name the taboos, we want something to laugh at.
>>16495183You ain't gonna laugh last, faggy fatty.https://www.youtube.com/watch?v=ebqAH5mLZNk
>>16495330is this a childhood friend of einstein?sorry, but where exactly is thinking about non-bigbang solutions a taboo?if you are unpleased that this isnt accepted by publishers, it could be because bigbang theories can explain plenty of observations.an alternative should be able to come up at least with a fraction of explanations, do you think?I would also be careful about accepting such publications, as this could have a negative impact on your reputation.is this the only 'taboo' you see?thats not a lot of taboos you see, not to say none.one could almost think its not about that at all, but rather about creating the impression that there are many taboos.you have more?
>>16495429> decided not to watch the video> shat a lot of babble, didn't forget to mention einstein for no good reason
>>16496464do you seriously think a single outlier would get published in nature? Do you know how much insane evidence they have to have across maybe 7 main figures each with 10 panels, then 20 supplemental figures, detailed methods, and getting past reviewer comments? It's those things alone that make the hypothesis worth pursuing further. >become famous>get a book deal>become richthis almost never happens in academia and just shows that you only follow pop scientists and your perceptions are from youtube videos
>>16499514Why wouldn't they publish EVERYTHING which comes their way? They don't have to waste paper on it, they would save many hours of back-and-forth correspondence. They could just mark every article that way, which would also give them an opportunity to change the marks up and down, depending on what further research shows. The whole system is so obsolete it's ridiculous to the levels at which I'm not even sure that I should offer them such modifications.
>>16442461What if the opposite of what he is saying is true and one in seven IS to be trusted, but that he doesn't want you to trust those in favor of the other six untrustworthy ones.
>>16494767>theory ladeness of evidenceyou stem sissies should take a look at this
>>16442512More than that, theres also the fake preprints to account for
I don't trust any science that can't produce a working product that can be commercialized.
>>16505585thats based, if it can't be converted into something someone is willing to pay for then its useless.
>>16505585science research grant funds should be managed by venture capitalists imagine the progress we'd have when only worthwhile projects got funded instead of wasting most of the funding on rich nepobabby's useless navel gazing
>>16507343Nobody is ever going to publish research thats of any tangible value. Valuable research is kept secret so the discoverers can capitalize on it themselves. An example is the case of the first person to figure out how to cultivate morels. Morels are normally a highly prized and expensive mushroom because they can only be collected in the wild, by in the 1980s a researcher at the University of San Francisco managed to grow some in the lab, he never published his research, instead he sold the idea to the owner/founder of the Dominoes Pizza chain, Tom Monaghan, who thought he could potentially sell a lot of pizzas by offering up previously expensive mushrooms cheap. Once Monaghan had the secret he had the researcher he bought it from assassinated, burnt down his lab and ransacked his house to make sure nobody else could ever get the information. Subsequently he decided that he didn't need to do the morel project to get rich and the lab method that was employed never saw the light of day.
>>16508425It's called a trade secret. Most research is useless without someone pouring millions or billions into figuring out how to exactly implement it in the real world and that implementation is kept secret, not the underlying idea underneath it.
I think about this every time I see acupuncture studies, which are often from china. An entire sub industry of doctors specialized in it is spawning off these studies.
>>16442461Finally, quantification of trust
>>16442512>7 in 7 are fakeEh, 5 of 7-Or , if you prefer:7 of 9!
>>16461470>You need to evaluate their methods and the data presented and decide if you agree or are convinced. If something is consistently evaluated properly, replicable, and observed...That is the PROBLEM!Peer Review isn't 'scientific' anymore.That is: If the work is in anyway contrary to standard 'Mainsteam Science', it does Not get published in any Major publication!Mainstream Science it BS!.It is tier-2 Science REAL Science is 'Classified' AKA - Not for public release!Pubic science is ' Disinformation '- Fake -
>>16510318>Peer Review isn't 'scientific' anymore.>anymoreit never was scientific. it was invented as a means of circumventing the scientific method
>>16511310Peer review is a form of the 'appeal to authority' fallacy
>>16445367I've obtained plenty of proof for strings, you faggot incel virgin AHAHAHAHAHAHAHHAHA
>>16442461can confirm, i edit a bunch of reatrdddd phd and post doc work. they’re all a bunch of liars.
>>16508425Mycologist Steven Hayden Pollock?Tell us more.
>>16442461is this because of dishonesty or incompetence?
>>16513411Both, they're fraudulent due to dishonesty and the fraud gets discovered due to incompetence. So 6/7 of papers not detected as fraudulent contain a large portion of undetected fraud. The fact that nobody ever reads most of those papers, not even the peer reviewers, means that most fraud if never detected.
>>16505585Engineerist here, this is the conclusion Ive come to as well
>>16515476playing useless sandbox games and pretending that you're doing something which constituted being a valuable member of society is an indicator of deeper mental illness
>>16516566>an indicator of deeper mental illnessAcademics get stuck in academia because they aren't capable of becoming valuable members of society who serve a legitimate purpose
>>16505585
>>16495159>hover over link to see the thumbnail>"Startup Health Now">"Human Longevity, Inc">hover away from the link
>>16442461>Despite making up only 20% of science papers, psychology makes up 50% of unreproducible papers.
>>16517956oncology has the highest rate of retractions, psychology is way, way down the list. retraction rate correlates with how much money there is in the field cause scientist are greedy fuckers who are become more prone to lying the closer they get to money
>>16505585quads of truth
>>16442512Niggers hate science. Niggers hate anything intelligent.
>>16518609Both are pretty high on the list (oncology is the worst), but psychology is pretty heavily skewed by social psychology (in psychology's defense).
>>16520354negroes are by far the most successful and important scientists
>>16521157Based negroes doing the heavy lifting in the sciences that other races refuse to do
>>16521157Science is a cult, move em to /x/
>>16520354They're too low IQ to know any better. Dumb people have dumb opinions and do dumb things. Expecting any better of them is insanity. >b-b-but dumb people should just be smarterThey can't, they don't have the brainpower to do so
>>16520354lmfao
>>16521157>Musk: TSMT gibes me more H1Bs so I can import 6 trillon of them
>>16478563>>16478697do they take turns cheating?
>>16526057one shits in the street while the other is 'publishing' elsewhere although even to the trained observer the two activities are often indistinguishable
>>16527230Shitting in the street is a form of publishing. Thats how animals mark their territory and circulate other information.
>one in seven
I think they put out fake studies to say seed oils are fine
>>16521157>if you look like a scientist you can show the children how to kiss a scientist in the mirror, but only on the lips
gee i sure do wonder why the time between posts is so long for this thread in particular
>>16531808https://en.wikipedia.org/wiki/Persecutory_delusion
>>1644246199% certain this is limited to psychology and medicine.
>>16533332it isn't, its everywhere, its probably worse in physics than it is in medicine
>>16534742Physics is a meme thats become entirely separated from reality at this point.
>>16442461if a peer reviewed paper says 1 in 7 are fake, then it's probably 5 in 7 in reality.
>>16442461>ONE IN SEVEN SCIENCE PAPERS FRAUDULENT>According to a new science paper:>One in seven science papers is not to be trusted, says new science paper
>>165360636 in 7 chance the paper is correct. Your reliance on a 14% likelihood of it being fake is a cringey desperate cope
I'm sure one of you has a collection of links regarding all these stats that have been mentioned here. Care to share?
>>16538083http://lemonparty.org
>>16442461Not surprising.My reviewers often only take a cursory look at my proofs and accept them if they "intuitively seem correct".For experiments, we have this thing called artifact evaluation in my field, which is to ensure the experiments are reproducible, but all they do is take our code, follow our instructions, and run it to see if they get similar results to ours.For all they know, the code could be completely bogus.Also, no reviewer reads your appendices
>>16442461>Interestingly enough all the submissions from Pfizer employees between 2020-2022 were found to be fake
>>16442461So in reality 1/8 papers
>>16533332
>>16442461whats the big deal, cant you just throw away every 7th submitted study ?what a nonproblem
>>16442461It's their fault for flooding graduate programs with cheap foreigners who have nothing to lose AND do not want to go back to their home country. Well permanently poisoned.At the very least we have stats on this. It's very easy to get things by as well as >>16540562 stated. I will also blame tech bugman for this all too. Guess what, like in every fucking field, we have humans working too fast, being told to work too fast, being too multidisciplinary, and having zero support offices because of fucking computers. Everything is lean even if science is 'wider' now. Another factor in our collapse is this replacement of humans by machines who are unfortunately not the same. They have plenty of benefits but overlords only see efficiency and cost saving, not nuance. Also the number of smarties hasn't gone up, but computationally demanding fields have...
>>16533332Look up Schön, he was a huge disaster in superconductivity Physics.>>16534742I sure hope not, life sciences is now so bad that it will be hard to top that, and were that to happen, it would render all papers useless.>>16535832Schön worked in a very applied field.
>>16538083The link to the paper is in the OP>>16442461The paper itself is attached to the second post>>16442462
>>16541370>it would render all papers useless.name anything useful that has come out of academic physics so far this century
>>16442669>science deniersredditoid
>>16481460Elaborate.
>>16442461>ONE IN SEVEN SCIENCE PAPERS FRAUDULENT>According to a new science paper
In unrelated news, fossil fuel and supplement companies found to pay for 1 in 7 science papers.
>>16466939>no sourceI don't know what was deleted but you're clearly retarded.
>>16543286Topological insulators. Wait, that was IBM and they never published.Neutrino detectors in the Antarctic... wait, that wasn't actually published either. Single photon spin sorting.. wait, also not published.Fugg
>>16545405>>no source>I don't know what was deletedthe source, retard. I don't know why you feel that you have to comment on something when you have no idea what it's about.
>>16442461Only 1 in 7? Hard to believe.
>>16545429nothing thats useful will ever get published, if someone discovers something useful they will keep it secret so that they can exploit it for their own personal gain
>>165474171 in 7 science papers is fraudulent 7 in 7 science papers is useless
>>16442461That must be one of the fake ones.
>>16549232ok but is it over or under exaggerating the fakeness factor in scientific publishing?
>>16553198Its isn't possible to overexaggerate he fakeness factor in that sector of the vanity publishing business
no
