/sci/ - Science & Math

[Anonymous No.15961546]

trust the science(TM) moment
https://twitter.com/LeahLibresco/status/1743769727330029639

[Anonymous No.15961548]

>error gets found with science
>somehow this is bad

[Anonymous No.15961554]

>>15961548
>"scientist" makes idiotic error
>proofreading misses it
>peer review misses it
>gets published
>picked up by large media outlets for clicks
>retracted months later when other scientists couldn't replicate the calculations
How about as a compromise we all agree to only Trust the Science until a year after publication (or in the case of politically sensitive topics, a year after it becomes culturally permissible to consider the possibility that the study was wrong)

[Anonymous No.15961558]

>>15961554
>proofreading misses it
>proofreading
What do you mean that the spellchecker missed a design flaw in the study? That is outrageous and shows that all of academia is fraud.

[Anonymous No.15961559]

>>15961554
everyone tells you replicated studies and meta analyses are the gold standard, but you just won't listen

[Anonymous No.15961561]

>>15961546
Social sciences in general are a meme and should be given the same level of consideration as astrology and psychology.

[Anonymous No.15961566]

>Shortly after the paper was published some colleagues from Bowling Green State, I-Fen Lin and Susan Brown, emailed me and my co-author about our estimate of divorce. They were trying to replicate the paper and couldn’t understand why their estimate was so much lower than ours. I sent them the statistical analysis file, which documents all of the steps as to how we came to all the estimates in the paper. And they pointed out to us, to our horror, that we had miscoded the dependent variable…As soon as we realized we made the mistake, we contacted the editor and told him what was happening, and said we made a mistake, we accept responsibility for it.
>The original code will be available online. The code that we used for all the estimates. That’s part of good research practice: somebody should be able to replicate your results. We talked with I-fen Lin and Susan Brown, and really appreciated them raising this with us. While you would never want to discover that you made a mistake, what’s ultimately important is to do good research, and sometimes that requires you to make a correction. We’ve tried to be completely transparent about the mistake that we made and correcting it as quickly and clearly as possible.
So, science corrected itself successfully. That's why I trust the science.

Also, the publication and retraction both happened in 2015. Why is this being pushed on social media now, in 2024?

[Anonymous No.15961570]

>>15961546
I think it's either a bot account or a mentally ill person. "She" posted 24 tweets in the last 2 days.

[Anonymous No.15961572]

>>15961566
no idea but how often do things like this get caught? and how far did the bad data get sensationalised by the media?

[Anonymous No.15961579]

>>15961572
>how often do things like this get caught?
Given that we're discussing a 2015 paper, I'd assume that these things don't happen so often.
>how far did the bad data get sensationalised by the media?
That's a media problem, not a problem of science. In my field we're often reliant on the latest publications. However, a single, not-yet replicated paper is not seen as proof for anything. We sometimes try to replicate the results ourselves and publish the results of that attempt. Sometimes we wate and only when other researchers come to the same conclusions, we start treating it seriously. Some IQ 80 sensationalist journalists going apeshit about such a result is really not my problem.

[Anonymous No.15961583]

>>15961554
>peer review misses it
you know how peer review works?
>oldfag head researcher or prof gets paper to review
>lazy, forwards it to some lackey PhD student
>PhD student skims it for some keywords, looks at the picture, quickly fills out the form

[Anonymous No.15961594]

>>15961579
>>15961566
Who gives a shit? The damage to public perception has been done. It's easy to cause and nearly impossible to fix. You can thump your holy journal in your ivory tower with your thumb up your ass about the triumphs of science, and it means absolutely Jack shit if public perception of science of consistently decreasing. The type of tone deaf self fellating going on in your posts is precisely the type of idiocy that is causing the public to lose faith in science. You know what I say? Deserved.

[Anonymous No.15961595]

>>15961594
>it means absolutely Jack shit if public perception of science of consistently decreasing
Why should I give a fuck about what "the public" thinks about my field? Do you think that bus drivers care what cosmologists think about bus drivers?
>The type of tone deaf self fellating going on in your posts is precisely the type of idiocy that is causing the public to lose faith in science.
You're confusing science and religion. Science doesn't rely on faith.

[Anonymous No.15961605]

>>15961595
>Why should I give a fuck about what "the public" thinks about my field?
Did I say "your field" you dumb kike, or did I say science in general? Holy shit you need to get your head out of your ass. If you really can't understand why publix perception matters then I reiterate: deserved.

[Anonymous No.15961933]

>>15961566
Because this happened and still hasn't been corrected.

[Anonymous No.15961938]

>>15961548
Strange how "errors" always seem to be in the direction of demonizing a particular demographic group. How many other "errors" demonizing this group that's trendy in academia to hate are still out there?

[Anonymous No.15961972]

>>15961595
Where do you get your funding?

[Anonymous No.15961995]

The Truth About Climate Change
Authors: Jonathan Tooker
https://vixra.org/abs/1309.0069
Climatology occupies the intersection of science policy and public understanding of science. In such a prominent position, the wide spectrum of climate opinions is remarkable. Society has achieved a paradigm in which global warming subscribers and non-subscribers are largely segregated by political affiliation. Since science is non-political, only a misunderstanding of the science can facilitate such a segregation. In the first section we analyze a recent study by Cook \emph{et al.} finding overwhelming scientific endorsement for the greenhouse theory of anthropogenic global warming (AGW). We find the popular reporting on Cook's result is not accurate. The aim of the following section is to clarify the science behind the most popular climate arguments and introduce the reader to some evidence that is not widely publicized. Even the astute non-climatologist should come away from this report with an enhanced understanding of relevant issues in modern climate science.

[Anonymous No.15962014]

>>15961570
>either a bot account or a mentally ill person
>either a bot account or a person who uses twitter
it’s the same proposition

[Anonymous No.15962056]

>>15961995
climate change is debunked by taking into account real numbers in the neighborhood of infinity in climate models

[Anonymous No.15962087]

>>15961554
Going by time is pointless. You can publish some shit that nobody cares about at the time and nobody feels the need to verify then it becomes relevant 10 years later.
Time can be a poor proxy for amount of scrutiny.
>>15961566
>So, science corrected itself successfully.
Only after someone else tried to replicate. All of the gatekeeping and quality control before the replication attempt were useless. The peer review system should have caught this. They were either too trusting to request the data or too lazy to catch the error.

[Anonymous No.15962163]

>>15962087
>Only after someone else tried to replicate
That someone else is also part of the science community.

[Anonymous No.15962175]

>>15962087
>The peer review system should have caught this
how?

[Anonymous No.15962205]

>>15962163
>That someone else is also part of the science community
Do you not know what peer review is? It is not simply just another scientist (who is a peer in some sense) reviewing your work by doing the experiment/study themself.
That is just science. Peer review is a sawed-off shortcut that aims to spot obvious errors/frauds just from reviewing the documents. The whole point of the process is to save time/money by catching the problems which don't require running the entire experiment again.
>>15962175
By requesting the same data and finding the same code error the other researches did.

[Anonymous No.15962299]

feminists used that study fucking constantly lol so I just assumed it was busted

[Anonymous No.15962303]

>>15961558
lmfao this nigger doesn't even know what proofreading is in an academic context.
>PROOF
hmmmm wonder what that might imply lmfao

[Anonymous No.15962306]

>>15962205
>finding the same code error the other researches did.
are you unironically illiterate? the error was found because other researchers failed to DUPLICATE the result, which made the original authors check their internal procedure again

[Anonymous No.15962425]

More proof that all scientific publishing is completely fake and contrived in order to to push a political agenda

[Anonymous No.15962434]

>>15961548
found the libtard

[Anonymous No.15962469]

>>15962306
>are you unironically illiterate? the error was found because other researchers failed to DUPLICATE the result
Are you an imbecile? The error was obviously found because that researcher's dad fucker her mom, which allowed her to be born, become a researcher, then fail to duplicate the results, then request the data.

We can play this game all the way back. I understand failure to duplicate is what created the motivation to look for an error. Since peer reviewers aren't in the business to redo entire studies, they should assume there might be errors from the start and try to find them.
The obvious necessary ingredient was for somebody to spot the code error. Spotting a code error should be something a peer reviewer does. If they don't think to request the code or they fail to vet the code then it is the peer reviewer's fault for signing off on something faulty.

[Anonymous No.15962583]

>>15961566
You think science has gotten *better* since 2015?

[Anonymous No.15962595]

>>15961579
>Given that we're discussing a 2015 paper, I'd assume that these things don't happen so often.
There's a whole website for this (Hint: It's linked in OP). Why this twitter rando unaffiliated with the site is interested in this specific case today is irrelevant.

Just for fun here's a running list they keep of 388 (so far) retracted chink flu papers:

https://retractionwatch.com/retracted-coronavirus-covid-19-papers/

[Anonymous No.15962609]

>>15961548
It's only bad because people will see "One single dubious social science study says" and make life choices based on it.

[Anonymous No.15962967]

>>15962595
>Just for fun here's a running list they keep of 388 (so far) retracted chink flu papers:
And how many were not retracted?

[Anonymous No.15963201]

>>15962967
>And what about all those people I *didn't* murder?

[Anonymous No.15963206]

>>15963201
>a methodological flaw in a scientific study that was discovered after publication is literally the same as murder

[Anonymous No.15963227]

>>15963206
>>15961933

[Anonymous No.15963250]

>>15963227
>read my wall or text and you'll see why scientists noticing flaws in their work and retracting a publication is the same as literal murder
Nope. Still not seeing it.

[Anonymous No.15963357]

>>15962469
>Spotting a code error should be something a peer reviewer does. If they don't think to request the code or they fail to vet the code then it is the peer reviewer's fault for signing off on something faulty.
Doesn't work like that. Reviewers are generally from the same field and more often than not don't know how to code, let alone review code.

[Anonymous No.15964992]

>>15962175
>peer review

[Anonymous No.15965561]

>>15963357
>Same field
>Reviews work of researcher who had some simple code to exclude certain data
>Somehow can't do/review the simple coding themself
For all I know it was just a formula in an excel sheet.
If you are in any science field and can't write a formula or code to exclude data based on a boolean value or flag then I find that hard to believe.

You mean to tell me every statistical analysis in peer reviewed papers isn't checked? It's all "just trust me, bro"?
Every graph/table comparing things by age, race, sex, etc. is just wizardry to the reviewer?
Even though, they have the same ability to make those graphs/tables in their own papers?
You are so full of shit.

[Anonymous No.15965643]

>>15961548
An obvious error got past peer review. This is evidence that peer review is unreliable.

[Anonymous No.15965724]

>>15965643
an obvious error didn't get past peer review. this is evidence that peer review is reliable

[Anonymous No.15965732]

>>15965643
If it was obvious, then the authors wouldn't have made it. If it was obvious, the peer reviewers would have found it. If it was obvious, the peers trying to replicate would've found it instantly. It was only found after some further investigation and code review. Isn't that proof that the mistake wasn't obvious at all? Peer review for journals doesn't check every line of code. I've never even sent code to a reviewer. They send you a list of questions and requests to change your manuscript.
>This section requires further explanation
>I am not convinced that the observed phenomena can be explained with just ...
If you don't manage to convince them, they reject you. If you can provide additional explanations that convince them, you pass. This isn't a pull request where a colleague reads every line of code.

[Anonymous No.15966034]

>>15965732
>If it was obvious, then the authors wouldn't have made it.
why not? most people publishing in the replication crisis press are low IQ, smart people don't devote their lives to vanity publishing

[Anonymous No.15967288]

>>15965732
>If it was obvious, then the authors wouldn't have made it.
Making mistakes is kinda independent of obviousness. Obviousness is a measure of how easy it is to spot.
>If it was obvious, the peer reviewers would have found it
You are assuming what needs to be proved. The argument is the peer reviewers should have spotted it because it would be obvious to anyone reviewing the data/statistical methods.
>If it was obvious, the peers trying to replicate would've found it instantly.
They did when they tried to crunch the numbers themselves. They immediately saw a disagreement and knew there had to be an error and where it would be in general (without having access to the code).
They notified the original researcher there might be an error and where it might be and it was instantly spotted in the code.

>Peer review for journals doesn't check every line of code. I've never even sent code to a reviewer.
So it is as useless as I am arguing. All data and statistical analysis is probably full of errors if that is the norm.
Publishing data and code should be the standard for all papers since peer review doesn't seek to spot errors there.
Wouldn't it be nice to be able to save time and money in the vetting of scientific knowledge by being able to spot errors without needing to have somebody else redo the entire experiment/study?
Why only vet the integrity of the methodology when flaws/fraud could just as easily hide in the data and data processing?

[Anonymous No.15967301]

>>15961546
What an obvious mistake. Men are biological ultraloyal gigasimps for sick girls.

[Anonymous No.15967787]

>>15967288
>>Peer review for journals doesn't check every line of code. I've never even sent code to a reviewer.
>So it is as useless as I am arguing.
REEEE WHY ISNT EVERY FIELD IN THE WORLD LIKE MINE
What are you even upset about? The error was found very quickly and the paper was retracted 9 years ago.

[Anonymous No.15968407]

>>15961546
Why are research groups like this?

[Anonymous No.15968412]

>>15968407
Like what? Why they check other people's work?

[Anonymous No.15968462]

>>15967787
>What are you even upset about? The error was found very quickly and the paper was retracted 9 years ago.
I'm thinking of all of the ones that aren't found but could be if the full documentation was provided.
Do you not see the waste of time/resources it could take to try to replicate entire studies in order to find errors?
It is more efficient to find the errors which can be found just by looking at documentation by looking at the documentation instead of doing the entire study again.

Even if you think it is too much work for the peer reviewers, there is no reason full documentation shouldn't be provided/published for anyone else to scrutinize.

[Anonymous No.15968475]

is someone unironically getting paid to shill in this thread? it's a such a bizarre overzealous defense of "Science" (TM) that I don't see from any of my STEM peers. i teach at a university, and i'm talking about biology, chemistry, mathematics, engineering. none of the faculty members in these disciplines (not even the woke, ardent anti-religion and anti-white) ones will shill "Science" as hard as the guy in this thread. i mostly see it from social science people and paid science "communicators" or PhD students who've never peer reviewed a paper

[Anonymous No.15968479]

>>15968462
> I'm thinking of all of the ones that aren't found but could be if the full documentation was provided.
I think this is called schizophrenia.

[Anonymous No.15968483]

>>15961548
>>15961558
>>15961570
Just give up, you run out of credibility.

[Anonymous No.15968484]

>>15968479
I guess demanding transparency is for schizos now?
Or you just have the delusion that nobody ever makes errors in data processing?

[Anonymous No.15968489]

>>15961548
Religion and ideology is never mistaken because it never acknowledges errors. /sci/ loses. /pol/ wins. When will eggheads learn?

[Anonymous No.15968490]

>>15968484
I don’t know the exact legal framework behind this, if this is EU policy or the policy of all my previous employers (universities and research institutes), but open access to publications AND data were always required for me. Granted, when this paper was published, I was still a student, but it could hardly be more transparent than it is in my experience.