science fraud is common place

Indeed but he isn't going to acknowledge this or my premise which doesn't have anything to do with what has already been discussed.
wrs, Everyone (except maybe justintime :)) knows that the number of papers redacted for fraud is not the same as the number of fraudulent papers.

The number of papers redacted for fraud is a lower limit for the number of fraudulent papers. That lower limit is very low (0.0096% in 2007).

The question then is what is the percentage of detection of fraudulent papers?
We can only estimate the but science has this requirement called replication. Fraudulent papers will be found when some other researcher tries to replicate the results. Thus that detection rate is quite high.
My estimate is greater then 90%. That makes the percentage of fraudulent papers in 2007 roughly 0.01%.

What is your estimate, wrs?
 
My premise is that fraud is commonplace in humanity.
No, wrs: It is your premise. It is your job to support it with evidence.
Start by defining "commonplace", i.e. put a number to it.
Show that the "fraud" in your assertion is the same as scientific fraud (basically faking results).
Show that the peer review process is "commonplace" in humanity.
Show that a demand for replication of results is "commonplace" in humanity.
After that you have a basis for comparing the "fraud" that is "commonplace" in humanity to scientific fraud that is present in scientific journals.
 
Hardly so. This is not what I claimed either. Go back and read what I said. Blutoski admitted to fraud being the cause in some cases. QED

I get the impression a lot of the participants are talking past each other.

We have good numbers for detected fraud in scientific publications. What baseline would you accept as a way to measure 'nonscientific' fraud?

A related question... in the most common situation where scientists have done good work in a corporate-sponsored lab, and the business only published submitted the results they liked to the FDA for product approval... would you consider this 'scientific' fraud or 'nonscientific' fraud? That's relevant because it represents almost all cases where the FDA has had to withdraw a product after approval due to fraud. Scientists prepared papers in good faith; a business cherry-picked one for publication. The paper would not be withdrawn because its findings are true. My opinion is that this is business fraud, but not scientific fraud.
 
Let me put this to you as actual rigour and scrutiny: You run your experiment with two people overlooking you and your measurements. Copies are taken of all the logs, all the code, all the data collection and held separately while you work on them. All versions of the code and data are also held so we know which version was run on what. The code is reviewed by people outside your department. An approval board sits to decide which method - implemented in which code version - is to be used. The final results are checked, data point by data point. And that's just the collection.
Wow - MCH, you have almost described a slightly loosely done clinical trial :jaw-dropp.
You missed out double-blinding, patient consent, ethical review and maybe some other stuff.
You added some irrelevant stuff about having "overlookers".

This is nothing to do with the subject of the thread which is peer review.
 
This could be quite amusing, I should be doing something else. I expect there to be some issues about 'lost information' - but is lost information to a small number of experts offset against 'found information' to a large number of non-experts? :-)

Before:

"Development is not a gradual but a saltatory process. A combination of qualitative changes in form and function—thresholds—creates boundaries between a succession of quantitative intervals—steps. Thresholds can be modified by an altered time of appearance of structures and functions (heterochrony), especially during early ontogeny, to form an operational basis for the prolongation of juvenile characters and adaptability into later ontogeny. Whereas such prolongation enables juvenilization in phylogeny, analogous principles may operate on a much shorter time scale to produce the r-selection-like altricial and the K-selection-like precocial trends in ontogeny. The inherited capacity to adjust constantly to the environment (heterochrony) selects for structural, biochemical and behavioral improvements. The tendency is toward the precocial but the way back to altricial forms is left open. The heterochronous adaptations in early ontogeny can “reverse” the vulnerable specialization, should the environment become less stable and/or the community less competitive. Juvenilization, capable of turning gerontomorphosis into paedomorphosis in evolution, and heterochronous shifts of character anlagen, capable of turning a precocial trend into an altricial trend in ontogeny are both part of the same biological process which operates during early ontogeny. "

After:

"Development (ontogeny) is not gradual but in sudden steps. Combinations of changes in form and function create the boundaries between quantitative steps. Altering the time that structures and functions appear (heterochrony) can change boundary thresholds, especially during early development, to prolong [keep?] juvenile characteristics and adaptability into later development. ..."

Um. I confess, the rest of it is lost to me after these glasses of wine. I lose.
 
I'd also like to point out that such supplements DO get tested. Anyone can do a search of threads in this forum to see such tests. These supplements are being analyzed scientifically, and all the issues with them have been discovered by scientists, not the likes of justintime or Maartenn100. What these arguments boil down to is, essentially, that the scientific process is too slow for the tastes of some people.

Agreed, although I mostly meant not necessarily prior to marketing, and in any case, it's not required by any regulatory body. Many are never tested (most are never tested... i'd guess 99.9% are untested) because they are citing some other published research as justification for form/function claim and claiming close enough similarity to have been proven safe/effective by proxy.





It works, but eventually, and these people don't like that last phrase.

There's probably more than one motive for these types of postings... there is antiscience, there is anti-government, there is anti-expertise... sometimes it's personal as there are a lot of people out there who were prescribed things like Vioxx and cultivate permanent mistrust of any safety claim regardless of their general attitude toward science, government, experts &c.
 
Wow - MCH, you have almost described a slightly loosely done clinical trial :jaw-dropp.
You missed out double-blinding, patient consent, ethical review and maybe some other stuff.
You added some irrelevant stuff about having "overlookers".

This is nothing to do with the subject of the thread which is peer review.
heh yes I know; I find it quite fun that even 'in science' we know how to do rigour and scrutiny, but then only do it in very specific disciplines, and yet people 'in science' claim that somehow the confidence we get from all that scrutiny and rigour in medicine somehow applies to all science. And also claim that medicine is not science.

[edit: it's not fun. It's depressing]
 
We can only estimate the but science has this requirement called replication. Fraudulent papers will be found when some other researcher tries to replicate the results. Thus that detection rate is quite high.
My estimate is greater then 90%. That makes the percentage of fraudulent papers in 2007 roughly 0.01%.
How many papers are replicated? If only a few are replicated, then the detection rate will not be high.
 
Here's what you said.
No, here is what I said.

When I see an agency like the FDA forced to withdraw a drug, I can only conclude there was fraud involved in the process of release. If not, then the FDA is incompetent and it's highly likely that there are a lot of drugs causing a lot more harm that the FDA hasn't found. If the FDA is in fact highly competent and the drug release process is as well, then it stands to reason that the only way a drug could get approved and subsequently withdrawn, is via fraud.

There is nothing in this sentence that can, to a reasonable person speaking the English language, be taken to mean that you are only talking about some cases of fraud. This sentence, read by a reasonable speaker of the English language, appears to argue that ALL cases of withdrawn FDA approval are due to fraud. The highlighted word in particular supports this assertion, but the complete lack of anything resembling discussions of any other options, and the lack of any caveats, equivocation, or proposed alternatives is sufficient to lead a reasonable speaker of the English language to that conclusion.

Then you should bother to read what comes after it as a clarification or limitation. Maybe it's poorly put together ( reversing the sentences would be better)but at the very least that should engender a request for clarification, not a full scale onslaught.


No one, least of all me, has ever argued that NO fraud occurs. I have merely argued that you have over-stated your case and that your understanding of the clinical trial system is flawed. The fact that you consider the appearance of previously-undetected detrimental side effects upon generael use as evidence of flaws in the reaserch further supports this conclusion.

Now if you go back to where I began this, you will see that I said that all it takes is one case of fraud to hurt a lot of people in this situation. That is how this came up in the first place.

Not ones in papers but stuff that actually got implemented and lied about, all the time. I keep thinking about medicines here. All it takes is one fraud to harm a huge number of people.
 
In what way are any personal details of the poster relevant? If it's about the person, not the argument, then it's ad-hom.

First off, as a courtesy, please put your parenthetical context explanations outside the box in which you quote me, to avoid attributing your explicatory words to me.

If you find my question, addressed to Justintime, to be an argumentum ad hominem, report it to the moderators (click on the little exclamation point, bottom left) and they will take it under advisement.
 
.. he's pointed at 'normal' dishonesty rates in 'normal' populations, and reasonably conjectured that scientists will follow similar rates.
That is really missing the point, MCH. wrs has not done anything like that. wrs made a general and vague comment without any numbers (no rates :eek:!). wrs applied this general comment to scientists.

What you and wrs are forgetting is that scientists are not the general public. They are under an extreme amount of inspection and they know this.
 
Last edited:
Are you, or are you not, claiming that we do not have sufficient data presented in this thread to discuss rates of fraud in science?
We have some data in this thread that we can use to discuss estimates of rates of fraud. Detection rates are not those data.

Have you, or have you not, presented the evidence best paraphrased as "Detection rates are only lower limits to actual rates of fraud"?
I don't really understand this. Detections are obviously (?) only the lower limits to actual. Assuming no false positives.

Have you, or have you not, defending the notion that scientists commit fraud at the same rates as other people?
I have proposed the concept and am waiting to see if it's not acceptable, and if not why not.

None of this has much to do with my points about the implications of unknown fraud rates. But that was a long time ago.
 
First off, as a courtesy, please put your parenthetical context explanations outside the box in which you quote me, to avoid attributing your explicatory words to me.
Fine - still learning how to cope without nested quotes.

If you find my question, addressed to Justintime, to be an argumentum ad hominem, report it to the moderators (click on the little exclamation point, bottom left) and they will take it under advisement.
That sounds a bit serious. Just tell me how a question about him personally rather than his argument is not an ad-hom.
 
What you and B]wrs[/B] are forgetting is that scientists are not the general public. They are under an extreme amount of inspection and they know this.
We are not forgetting it. We (well, I) just don't think that researchers are significantly different from the general public.

I don't think it's unreasonable to start with the assumption that a researcher is bog-standard human, and then work out from there using, say, data.

I still want to know what these 'extreme amount of inspection' is. Because frankly, from what I've seen, academic researchers have considerably *less* scrutiny than many other areas.

As above, medicine is one area which *does* have an extreme amount of inspection - and the reason we have it is because medical researchers are like ordinary people and make mistakes, commit fraud, etc. What keeps them honest is the inspection. Where the inspection is missing - in other disciplines - we can expect that to rise.
 
I think the extent of fraud is important to the OP's question.
MCH: Read the OP again. There is no question. It is merely the assertion that "science fraud is common place" (with citation of a YouTube video :eye-poppi).
The content of the OP is a lot of myths and even paranoia about science research.

The fact is that scientific frauds - such as removing 'difficult' data points, or only looking for problems with unexpected results, or abandoning an experiment that appears to be 'going wrong' - will be discovered when the results are replicated and
* 'difficult' data points are seen
* unexpected results are seen and noted as problems
* the experiment goes wrong

Normally, we have quality controls in place to try and mitigate these ordinary human faults, and academia is not en exception (unless someone is ignorant about peer review, replication, community review).
 
wow, this sounds so wonderful. Let's all join hands ... .......
wow, wrs, that sounds so ... absurdly sarcastic and completely missing the point :eek:!

You do know that a requirement for empirical data to be generally accepted in science is that it is independently replicated?
 
None of those things you give are quality controls. Neither peer review, replication, or community review. What qualities exactly are they supposed to assure?

How often are results replicated? Several people have commented above that replication is sometimes impossible, or too expensive, or cannot get funding, or is not interesting enough to do, etc, etc.

Peer review is for interest, not for quality.

You cannot start quality controls at publication - it has to start at data collection.
 
wrs, Everyone (except maybe justintime :)) knows that the number of papers redacted for fraud is not the same as the number of fraudulent papers.

The number of papers redacted for fraud is a lower limit for the number of fraudulent papers. That lower limit is very low (0.0096% in 2007).

The question then is what is the percentage of detection of fraudulent papers?
We can only estimate the but science has this requirement called replication. Fraudulent papers will be found when some other researcher tries to replicate the results. Thus that detection rate is quite high.
My estimate is greater then 90%. That makes the percentage of fraudulent papers in 2007 roughly 0.01%.

What is your estimate, wrs?

I was not really interested in the fraudulent paper kerfluffle. I am more interested in this

Not ones in papers but stuff that actually got implemented and lied about, all the time. I keep thinking about medicines here. All it takes is one fraud to harm a huge number of people.

My initial premise is that fraud is commonplace in humanity. I base this on the fact that fraud is illegal. Therefore it must arise enough to be noticeable and thus a problem. We were discussing noticeable earlier in the thread. I said no higher than 6% maybe as low as 1-2%. Scientists are human and therefore fraud should be just as commonplace.

Papers are not the only place fraud is committed in the name or practice of "science". Any kind of scientific testing or process qualifies, especially if the public is relying upon it's claims. This makes the range of possibilities quite large, enough to be noticeable. Any arguments to the contrary in efforts to minimize the extent of "science" such as Perpetual Student made here are to my mind nothing more than the No True Scotsman ruse.
 
I'm amused that justintime, when looking for an example of scientific mambo jimbo, happened to choose

(a) a 1980-ish abstract by this particular author, who is obscure by most standards but (being a vaguely anti-Darwin contrarian) is frequently quoted by creationists. Did you find this by browsing scientific literature, or creationist literature?

(b) An abstract, in particular. Abstracts are the shortest and densest writing that scientists do. Why complain? Do you complain about the use of abbreviations in Twitter feeds, or digit codes on police radio ("207 in progress at Pico and 3rd")? Is the baseball box score an attempt to hide something about the game?

Cardinals 0 0 0 0 0 0 1 0 0 1 9 1
Red Sox 0 0 3 3 0 0 0 0 X 6 8 1

2B: S Victorino (2, off M Wacha); J Ellsbury (4, off M Wacha).
HR: S Drew (1, off M Wacha; 4th inn, 0 on, 0 outs to Deep CF-RF).
IBB: D Ortiz 3 (4, 2 by M Wacha, by T Rosenthal).
HBP: J Gomes (1, by M Wacha).
TB: S Drew 5; S Victorino 3; J Ellsbury 3; J Gomes; M Napoli.
RBI: S Victorino 4 (12); S Drew (4); M Napoli (7).
2-out RBI: S Victorino 4; M Napoli.
Team LOB: 11.
With RISP: 3 for 12.
Fielding
DP: 1. S Drew-M Napoli.
E: D Pedroia (1).

That's Greek to the uninitiated.

(c) You made a big mistake in picking this test case. You didn't read the article, did you? You stopped at the abstract. In this case, not only is the article itself fairly readable---the article goes out of its way to be even more readable by including a glossary.

This paradox applies to parturition and hatching (see Glossary on p. 589 for definition of this and other terms used in this paper), both of which are impossible to define in terms of "normality" because both are largely influenced by the environment and do not occur at a particular state of de- velopment.

...

GLOSSARY
Gene Helfman complained that "much of the con- troversy surrounding the general acceptance of [my] developmental hypotheses stems from difficulties with terminology." His request for a glossary was repeated by Gregory Whitt and others. While most of the terms are explained in the text and are well defined in my earlier papers, I feel that compliance with this request is justified, if somewhat at variance with my Popperian altitudes. Having made a valiant effort, I nevertheless feel that most of the brief definitions are far from adequate due to the complexity of the concept or because the content can only be fully comprehensible in a historical context. I am not interested in contributing to the debate about terminology. I am, as I was in the past, forced to address that subject in order to clarify what I mean in a conceptual context. Adaptations of old terms proved impossible due to their past association and wrong usage.

HETEROCHRONY: altered time of appearance of struc- tures and functions caused by varied strength of internal factors of development exerting their ef- fect at different rates, i.e., shifting or "changing of gears" along a time scale.

etc.

That's a rather stark contradiction (to say the least) to your claim that this author is deliberately mumbo-jumboing as a form of fraud.
 
Alright, it's the economist, but:

http://www.economist.com/news/brief...elf-correcting-alarming-degree-it-not-trouble

" A few years ago scientists at Amgen, an American drug company, tried to replicate 53 studies that they considered landmarks in the basic science of cancer, often co-operating closely with the original researchers to ensure that their experimental technique matched the one used first time round.

According to a piece they wrote last year in Nature, a leading scientific journal, they were able to reproduce the original results in just six.

Months earlier Florian Prinz and his colleagues at Bayer HealthCare, a German pharmaceutical giant, reported in Nature Reviews Drug Discovery, a sister journal, that they had successfully reproduced the published results in just a quarter of 67 seminal studies.
 

ISF - Join now!

Every member here is approved by hand. No bots, no spam, just people who care about evidence and honest debate.

Membership is free!

Create your free account

Back
Top Bottom