science fraud is common place

New set of data to establish science fraud is common place.

Unreliable Research Trouble in the Lab.
Quotes from link that question fraud, self correcting non existent, non replicable research, only positive statistics are published, fraud goes under-reported and scientists have a poor understanding of statistics.

1. When an official at America’s National Institutes of Health (NIH) reckons, despairingly, that researchers would find it hard to reproduce at least three-quarters of all published biomedical findings, the public part of the process seems to have failed.

2. Evidence that many more dodgy results are published than are subsequently corrected or withdrawn calls that much-vaunted capacity for self-correction into question. There are errors in a lot more of the scientific papers being published, written about and acted on than anyone would normally suppose, or like to think.

3. “There is no cost to getting things wrong,” says Brian Nosek, a psychologist at the University of Virginia who has taken an interest in his discipline’s persistent errors. “The cost is not getting them published.”

4. If the likelihood of such a false-positive conclusion is less than 5%, they deem the evidence that the hypothesis is true “statistically significant”. They are thus accepting that one result in 20 will be falsely positive—but one in 20 seems a satisfactorily low rate.

5. But most scientists are not statisticians. Victoria Stodden, a statistician at Columbia, speaks for many in her trade when she says that scientists’ grasp of statistics has not kept pace with the development of complex mathematical techniques for crunching data. Some scientists use inappropriate techniques because those are the ones they feel comfortable with; others latch on to new ones without understanding their subtleties. Some just rely on the methods built into their software, even if they don’t understand them.

6. Fraud is very likely second to incompetence in generating erroneous results, though it is hard to tell for certain. Dr Fanelli has looked at 21 different surveys of academics (mostly in the biomedical sciences but also in civil engineering, chemistry and economics) carried out between 1987 and 2008. Only 2% of respondents admitted falsifying or fabricating data, but 28% of respondents claimed to know of colleagues who engaged in questionable research practices.

7. And scientists themselves, Dr Alberts insisted, “need to develop a value system where simply moving on from one’s mistakes without publicly acknowledging them severely damages, rather than protects, a scientific reputation.” This will not be easy. But if science is to stay on its tracks, and be worthy of the trust so widely invested in it, it may be necessary

8. Most academic researchers would rather spend time on work that is more likely to enhance their careers. This is especially true of junior researchers, who are aware that overzealous replication can be seen as an implicit challenge to authority. Often, only people with an axe to grind pursue replications with vigour—a state of affairs which makes people wary of having their work replicated.

9. ohn Bohannon, a biologist at Harvard, recently submitted a pseudonymous paper on the effects of a chemical derived from lichen on cancer cells to 304 journals describing themselves as using peer review. An unusual move; but it was an unusual paper, concocted wholesale and stuffed with clangers in study design, analysis and interpretation of results. Receiving this dog’s dinner from a fictitious researcher at a made up university, 157 of the journals accepted it for publication.
British Medical Journal, sent an article containing eight deliberate mistakes in study design, analysis and interpretation to more than 200 of the BMJ’s regular reviewers. Not one picked out all the mistakes.

10. As well as not spotting things they ought to spot, there is a lot that peer reviewers do not even try to check. They do not typically re-analyse the data presented from scratch, contenting themselves with a sense that the authors’ analysis is properly conceived. And they cannot be expected to spot deliberate falsifications if they are carried out with a modicum of subtlety.
 
The 14% number would only work if every scientist polled knew only one other scientist. If each scientist knew 10 other scientists, then the number would go down to 1.4%.
 
If maths is a matter of opinion then I demand my A level is given a grade that reflects the answers I gave were correct in my opinion.
And I would like to announce here and now that I actually got an A in calculus instead of the D I received only because of a pretty steep curve.


Billions are spend storing useless pieces of electronic conversations or archiving trash talk that have no socially redeeming value except to remind us how we wasted our time in mindless social networking with the least possibility all this effort would amount to anything.
Well, you could stop your part in this right now. Quit reading and posting, turn off your computer, and go out and tell the world. (No bullhorns, PA systems or printing handouts or painting signs though - that would be using technology.) You could find yourself leading a movement. :rolleyes:
 
If computers and other digital gadgets had any practical application, why do you see millions of them dumped annually in landfills like useless garbage or disposable waste?

Because, like every other built object, they become obsolete or break down to the point of costing less to replace than to repair. Perpetuity is not the sine qua non of utility.

All computers have done is afforded us an efficient way to waste our lives.

I beg to differ. But we've already been through this part of the discussion. You happily use a computer and you happily seem to enjoy all the things we do better and more efficiently with computers. Granted, we're using computers only as one of many possible examples of the way science benefits humankind. But as with all such examples, if this is what you really believe, then eschew their use and go back to the medieval way of life. The fact that you don't indicates you don't really believe your argument has merit.

But because we are using computers to communicate(eletronically), many feel obligated to give science fraud a pass.

No one is "giving science fraud a pass." We're pointing out that, contrary to your invented and inflated numbers, it does not occur at a rate that merits hysteria.

We point out that you happily use the products of science and industry as a way of rebutting your moralist argument thinly veiled in the notion that nothing good can come of it. Hypocrisy has a way of dulling the edge of moralism.

When is science going to be used for something practical that makes people want to end science fraud instead of watching scientists slowly destroying our humanity.

Asked and answered. Numerous examples have already been given, which you either ignore entirely or attribute to some sort of spontaneous genesis.

You have not proven that science fraud occurs in any but the tiniest amount. Please don't continue to argue as if your principal proposition is a foregone conclusion. That's just begging the question.
 
6. Fraud is very likely second to incompetence in generating erroneous results, though it is hard to tell for certain. Dr Fanelli has looked at 21 different surveys of academics (mostly in the biomedical sciences but also in civil engineering, chemistry and economics) carried out between 1987 and 2008. Only 2% of respondents admitted falsifying or fabricating data, but 28% of respondents claimed to know of colleagues who engaged in questionable research practices.

How the hell did 14% become 28 ?
 
Why did you fail to provide the actual link so that we could read the article ourselves and draw our own conclusions? Why did you just cherry-pick quotes from it?

Technically speaking, that post was an infraction--you're not supposed to simply copy and past large swaths of other people's work. But since this is actually comparatively honest for justintime, I for one am willing to give it a pass.

Most of what that rant discusses is not fraud. For example, a 1-in-20 chance of being wrong (p=0.05) is certainly not fraud--it's merely setting limits to what one will accept as significant. If JiT thinks that's bad, he should see some paleontology work--we routinely use p=0.1 and higher! The issue is that biological systems are variable, and you can't demand more accuracy than the variability of the system allows for, or than the data can support. You can calculate to the nineth decimal place variance between ostrocods--but if your dataset is five specimens, it's rather disengenuous.

Furthermore, it mischaracterizes the analysis of these reports. The analyses always acknowledge that they are working with a variable population, and are making at minimum a good-faith effort to describe that variation accurately and appropriately. Such good-faith efforts may fail; just because you attempt something doesn't mean you succeed. But to call a good-faith effort fraud merely because it fails is beyond the bounds of civil discussion, and enters into slander and libel.

The idea that science isn't self-correcting has been discussed to death, but it apparently bears repeating: science is self-correcting over time. The correction need not come within the same generation--it may happen a century from now. It depends on when researchers will consider the question interesting enough to spend their limited time, effort, and resources on. But the important point is that correction will come--and it comes from scientists. In tens of pages justintime has not once provided a single example of a non-scientist discovering errors in science. That alone proves the efficacy of science as a self-correcting process: if it were less effective, justintime would be unable to point to its ineffficiencies.

The rest of the post is of similar quality. It has a window-dressing of scientific understanding, but that's all. There's no depth to that understanding in that post.
 
The 14% number would only work if every scientist polled knew only one other scientist. If each scientist knew 10 other scientists, then the number would go down to 1.4%.

The new study raises that number to 28%.

" Only 2% of respondents admitted falsifying or fabricating data, but 28% of respondents claimed to know of colleagues who engaged in questionable research practices."
http://www.internationalskeptics.com/forums/showpost.php?p=9784294&postcount=1322
 
1. When an official at America’s National Institutes of Health...

The Economist does not name the source nor give any basis for the estimate.

2. Evidence that many more dodgy results...

The Economist provides no evidence for this assertion.

3. “There is no cost to getting things wrong,” says Brian Nosek, a psychologist at the University of Virginia who has taken an interest in his discipline’s persistent errors. “The cost is not getting them published.”

Note the quote from Albert Einstein on Dr. Nosek's home page. http://projectimplicit.net/nosek/ I don't think Nosek agrees with you that science is worthless. What this means is that we have to put the cherry-picked quote the Economist author chose to use in the context of all this person's work, as we will do with similar quotes in the article.

Dr. Nosek's bibliography is here http://projectimplicit.net/nosek/papers/ . The relevant paper is arxiv.org/pdf/1205.4251 .

Oh look! It's a scientist noting what he believes to be a problem in science and is taking steps within the scientific community to address it. Is this behavior you would expect from a group based on fraud?

4. If the likelihood of such a false-positive conclusion...

Why are you posting this ostensibly as support of science fraud? You already accepted the premise that proper statistical handling of empirical data might require careful attention to measures of central tendency and trimming the variance by discarding outlying data. Why are you now applying a double standard?

5. But most scientists are not statisticians. Victoria Stodden, a statistician at Columbia, speaks for many in her trade when she says that scientists’ grasp of statistics has not kept pace...

No basis given. While we can stipulate that Stodden is a qualified statistician, The Economist article does not say how Stodden arrived at her conclusion, nor reports any qualitative or quantitative basis for it. In reviewing her research interests and publications http://www.stanford.edu/~vcs/ , I note that she works with computation and statistics, but I fail to see where she has studied and acquired any knowledge about how other scientists in other fields are misusing statistics.

6. Fraud is very likely second to incompetence in generating erroneous results, though it is hard to tell for certain.

An opinion of The Economist, with no factual basis given other than the 2% study we've already discussed at length.

7. And scientists themselves, Dr Alberts insisted, “need to develop a value system..."

Probably a valid observation. But you spin it as scientists somehow knowing they're wrong or knowing they're producing irrelevant, false, or useless results rather than examples of an ongoing effort to maintain standards in scientific research. Every so often, even well-meaning authors pick up on this ongoing effort and write sensationalist articles in the popular press that promote an alarmist view.

What you're missing is that these efforts inevitably vindicate science as a self-correcting field, contrary to the authors' desired intent. These authors cannot help but support their thesis with evidence from within science that scientists are conscientious about improving their credibility and reliability. This is why it's always problematic to try to support one's arguments about science by quoting the popular press.

8. Most academic researchers would rather spend time on work that is more likely to enhance their careers.

The Economist provides no source for this claim.

9. John Bohannon, a biologist at Harvard...

Asked and answered. We covered this singular effort earlier in the thread.

10. As well as not spotting things they ought to spot, there is a lot that peer reviewers do not even try to check. They do not typically re-analyse the data presented from scratch...

The Economist merely asserts without justification that this is or should be a task of peer review, in a sense misrepresenting what peer review is meant to be. Peer reviewers are asked to comment upon the methodological validity of the study, which can be done by inspection of the paper. If the validity cannot be judged from the paper, then the paper fails the reproducibility criterion and should be rejected.
 
<snip>

If you ask Dr Math on the Math Forum if you can have greater that 100 percent. He says:

"There is nothing intrinsically wrong with using percentages greater
than 100%.

A percentage is simply another way of writing a fraction with a
denominator of 100. For example, 6% = 6/100. Just as you can have an
improper fraction (a fraction whose numerator is greater than the
denominator), such as 4/3, so you can have what we might call an
"improper percentage" like 120% or 300%.

Now, what if the number of questions received went up to 14000 in
April? (It didn't.) This would be an increase of 122% from March to
April. There is nothing wrong with this - no law says that the number
of questions can't do more than double from one month to the next.

So the answer is, there is such a thing as a percentage larger than
100%"


Edited by Loss Leader: 
Edited in accordance with Rule 12 edit above.
 
Last edited by a moderator:
Most of what that rant discusses is not fraud.

Agreed, but Justintime has gradually relaxed his definition of what he considers "fraud." Fraud, in his definition, now includes errors committed without malice. It also includes failure to self-correct to some arbitrary standard within some arbitrary time.

The idea that science isn't self-correcting has been discussed to death, but it apparently bears repeating: science is self-correcting over time.

Newton even corrected some of his theories within his lifetime, as he was able to measure certain physical phenomena more accurately. Subsequently we've had to correct Newton in similar fashion over time, according to the same root cause and procedure. But Newton stood for hundreds of years (and parts of it still stand) before his notion of dynamics had to be vastly revised.

The rest of the post is of similar quality. It has a window-dressing of scientific understanding, but that's all. There's no depth to that understanding in that post.

Correct. It's basically a sensationalist hit-and-run article that ignores a great deal of what science actually is about.
 
If you ask Dr Math on the Math Forum if you can have greater that 100 percent. He says:

Asked and answered. He goes on to say that in other cases having percentages greater than 100 makes no sense. We have shown you how this is one of those cases.

ETA: I though we had gotten past this. Earlier you seemed content with the notion that an over-unity result was necessarily erroneous. This is apparently why you labored so hard to contrive some math that gave you a result less than 100%. Why have we regressed to claiming over-unity is okay?
 
Last edited:
The new study raises that number to 28%.

No reference given to an actual study, just to an article in the popular press. In the same paragraph a reference to "Dr. [Daniele] Fanelli" is made, but the article does not connect her with the alleged finding.

You don't get to claim this figure and insinuate that it has been properly arrived and and reported until you know the actual science that produced it.

Sorry, but no dice.
 
Last edited:
Here is the only paper by Dr. Fanelli that purports to investigate, by means of meta-research, how many scientists have committed acts of scientific misrepresentation (<2%) and how many have knowledge of others' misrepresentation (~14%)

http://www.research.ed.ac.uk/portal...ta(9819a91d-3a9f-44eb-ba3f-64a3c99fd4f6).html

There is no basis here or in any other work by Dr. Fanelli, that supports a figure of 28% either as a perpetrator or as a witness. Since "Dr. Fanelli" is the only researcher mentioned in The Economist article in the vicinity of the claim, I reject that figure as unsubstantiated.
 
JayUtah said:
Newton even corrected some of his theories within his lifetime...
I certainly didn't mean to imply that this never happens. Many scientists correct their own work, and later publications include revisions to the findings of earlier publications. Other times, researchers will find that they are working on the same or similar studies, and will correct each other in print (this ranges from a friendly "My results show this, can you help me figure out what's going on?" to a not so friendly "My esteemed colleague's conclusions are nonsense and clearly contradict this data").

My only point was that there is nowhere in science any demand that self-correction be done quickly. It can happen quite rapidly--but it can also take a long, long time. Fundamentally, this is irrelevant to science; that it gets corrected is vital, but the timing is irrelevant. The reason is simple: science must allow its conclusions to be updated upon the discovery of new data, in perpetuity. NO timelimit can be enforced; to enforce a time limit creates dogma and unchallangeable assertions, which undercuts the very foundation of the grand experiment that is scientific investigation.

Agreed, but Justintime has gradually relaxed his definition of what he considers "fraud."
True. I just felt the need to point it out. Yet again. Because, you know, the first fifty times were so effective. :D
 
I certainly didn't mean to imply that this never happens. Many scientists correct their own work, and later publications include revisions to the findings of earlier publications. Other times, researchers will find that they are working on the same or similar studies, and will correct each other in print (this ranges from a friendly "My results show this, can you help me figure out what's going on?" to a not so friendly "My esteemed colleague's conclusions are nonsense and clearly contradict this data").

My only point was that there is nowhere in science any demand that self-correction be done quickly. It can happen quite rapidly--but it can also take a long, long time. Fundamentally, this is irrelevant to science; that it gets corrected is vital, but the timing is irrelevant. The reason is simple: science must allow its conclusions to be updated upon the discovery of new data, in perpetuity. NO timelimit can be enforced; to enforce a time limit creates dogma and unchallangeable assertions, which undercuts the very foundation of the grand experiment that is scientific investigation.

Quote:
Agreed, but Justintime has gradually relaxed his definition of what he considers "fraud."

True. I just felt the need to point it out. Yet again. Because, you know, the first fifty times were so effective. :D

I have not changed or relaxed my definition of fraud. Fraud is fraud and science fraud is increasing and becoming more common place.
The reasons offered for this increase in science fraud it the flawed peer review process and the self correcting belief in science is a myth.
 

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