science fraud is common place

Interestingly the article comments
But some new journals—PLoS One, published by the not-for-profit Public Library of Science, was the pioneer—make a point of being less picky. These “minimal-threshold” journals, which are online-only, seek to publish as much science as possible, rather than to pick out the best. They thus ask their peer reviewers only if a paper is methodologically sound. Remarkably, almost half the submissions to PLoS One are rejected for failing to clear that seemingly low bar.
PLoS One is where the survey meta analysis was published.

It is not remarkable that an on-line journal with a low threshold of peer review gets a lot of low quality papers submitted. These papers were probably first rejected by journals with higher standards.
 
Interestingly the article comments

PLoS One is where the survey meta analysis was published.

It is not remarkable that an on-line journal with a low threshold of peer review gets a lot of low quality papers submitted. These papers were probably first rejected by journals with higher standards.

And not surprising. It's notoriously difficult to get a meaningful, accurate confidence interval in meta-analysis. While there is language in the paper, and occasionally in the treatments of the paper, to that effect, it seems to be getting lost in the noise of debate. But the difficulty in determining confidence would be a show-stopper for many of the respectable journals I know of.
 
Did you understand the point ? Let me state it again:

Let's say that you live on a street with 100 people. I poll those 100 people and 14% of them (14, in this case) tell me that they know someone called Justin on their street. By your logic, this means there are 14 people named Justin on the street. But that's wrong. In reality there is ONE person called Justin, you, and 14 people know you.

It's the same scenario with the scientists. 14% of them report knowing someone who MAY have been involved in fraud. It does not follow that another 14% of them are frauds, and it certainly doesn't follow even from that that 14% of all papers are frauds, because that would assume that all papers by all scientists who have ever commited fraud are fraudulent, which certainly makes no sense.

Justin, you really need to address this.
 
Guys, I am pretty sure it is impossible to feed a troll to death.

True, but it IS possible to use them for our own amusement.

Reality Check said:
It is not remarkable that an on-line journal with a low threshold of peer review gets a lot of low quality papers submitted.
I do want to emphasize that not all online journals have low thresholds for peer review. Paleontologia Electronica, my go-to example of an online journal, is open-access, online-only, etc. That said, they are also a well-respected journal with high standards for their authors and reviewers. It's not a journal with an enormous impact factor, but it's a step above your workaday museum memoires publication (which are still highly respected accademic publications). My point is, I don't want people reading this confusing "online journal" with "crackpottery". Many of them are nonsense, yes, but there are those that are serious journals attempting to figure out this new publishing paradigm.
 
Agreed. Whether it's a rate of malfeasance or voodoo mathematics to make the numbers come out grossly exaggerated -- which is what Justintime admits he's trying to do -- it's a "fraudulent" finding as he defines it for others.

In other words, no moralist argument he might wish to make can fly here now.

This old chestnut seems to apply.

The problem isn't that figures lie but that liars figure.
 
Originally Posted by Belz... View Post
Did you understand the point ? Let me state it again:

Let's say that you live on a street with 100 people. I poll those 100 people and 14% of them (14, in this case) tell me that they know someone called Justin on their street. By your logic, this means there are 14 people named Justin on the street. But that's wrong. In reality there is ONE person called Justin, you, and 14 people know you.

It's the same scenario with the scientists. 14% of them report knowing someone who MAY have been involved in fraud. It does not follow that another 14% of them are frauds, and it certainly doesn't follow even from that that 14% of all papers are frauds, because that would assume that all papers by all scientists who have ever commited fraud are fraudulent, which certainly makes no sense.


Justin, you really need to address this.

Surveys are taken across a wide cross-section of the population and geography to give an accurate picture that reflects the sentiments of the people being surveyed on the topic under consideration.

The survey of scientists were not based on a group of scientists from a particular community or location or institution. It was of scientists across the country from all the various disciplines, organizations, institutions. The likelihood they all knew each other would have made the survey meaningless. Surely you must know candidates are picked at random in surveys and every effort is made to avoid collusion between candidates to eliminate the one to many and the many to one relationship. This invalidates your assumptions/example.

The survey found 2% of the scientists admitted to fraud. When asked about the behavior of their colleagues (people they worked with or known to them)14% 28% admitted they knew someone guilty of fraud but undetected.

What the survey revealed was when scientists are asked to admit their own fraud many are reluctant (for obvious reasons--self incriminating). But when asked to report on the behavior of others they are more willing to admit fraud is committed by more scientists than reported. 28% of scientists came forward admitting they knew someone guilty of fraud which is more that the 2% admitting to fraud and that backs the surveys conclusion more scientists commit fraud than is reported. How many more?

According to Betrayers of the Truth [William Broad, Nicholas Wade] for every major fraud a 100 go undetected/unreported.

But we already know 2% of scientists admit to fraud. So 2% x 100 = OMG! 200% of science is fraudulent!!!

And just when you think that is bad.

28% of scientists report they know of at least one other scientists guilty of fraud. Given that surveys are careful to avoid a one to many and a many to one relationship between candidates picked. A one to one is used in our calculation which yields 28% of scientists are actually committing fraud.
If so then according to Betrayers of the Truth [William Broad, Nicholas Wade] for every major fraud a 100 go undetected/unreported.
28% x 100 = OMG! 2,800% of science is fraudulent.

The only way these percentages will go down is if all the scientists committing fraud start admitting to committing fraud which we know will never happen. But if we can prove the percentage of past fraud drives these numbers even higher. We can get the public to accept science fraud is more than common place. It is in the OMG! range.
 
justintime said:
Surveys are taken attempt to sample across a wide cross-section of the population and geography to give an accurate picture that reflects the sentiments of the people being surveyed on the topic under consideration.
Subtle difference there.

The likelihood they all knew each other would have made the survey meaningless.
The fact that scientists do know more than one other scientist thus renders the survey meaningless, by your own admission.

Surely you must know candidates are picked at random in surveys and every effort is made to avoid collusion between candidates to eliminate the one to many and the many to one relationship. This invalidates your assumptions/example.
No, it merely demonstrates that you have no idea what the criticism was.

Plus there are all the other known issues with surveys--such as how the questions were asked, who was defining the various terms, when the questions were asked, etc. Personally I consider surveys to be more or less of entertainment value only.

The survey found 2% of the scientists admitted to fraud. When asked about the behavior of their colleagues (people they worked with or known to them)14% 28% admitted they knew someone guilty of fraud but undetected.
The dishonesty of this statement has been shown so thoroughly that this "correction" can only be taken as an admission of guilt.

Then there's the fact that you apparently can't tell the difference between misconduct and fraud.

What the survey revealed was...
...nothing. At least, not to anyone familiar with how surveys are conducted or with interpreting the results of surveys.

According to Betrayers of the Truth [William Broad, Nicholas Wade] for every major fraud a 100 go undetected/unreported.
First off, that book has been called into question numerous times and should not be used as the Word of God the way you do. Second, YOU demonstrated that that number was pure speculation on the part of the authors. They made it up, and admitted as much. There is no reason to believe this number has any relationship with the facts of the matter being discussed.

But we already know 2% of scientists admit to fraud. So 2% x 100 = OMG! 200% of science is fraudulent!!!
Again, you are saying that 1) every case of detected fraud directly and immediately causes 100 other cases, unrelated to the original paper; and 2) for ever paper published, two are fraudulant. This is magical thinking at its worst.

The only way these percentages will go down is if all the scientists committing fraud start admitting to committing fraud which we know will never happen.
Funny how after several weeks of this nonsense, you still can't show a single example of a non-scientist discovering fraud in science. Every example has been scientists calling out other scientists for fraud. So we demonstrably are admitting that fraud occurs, and are policing ourselves. We can debate the efficacy of that process, but to say it doesn't happen is to deny the validity of your own evidence.

We can get the public to accept science fraud is more than common place. It is in the OMG! range.
0.005% is not in the OMG! range by any rational standard (I realize there's an assumption in there that's not applicable to you, however).
 
The survey of scientists were not based on a group of scientists from a particular community or location or institution. It was of scientists across the country from all the various disciplines, organizations, institutions.

No, it was explicitly the opposite. You didn't read the paper.

The survey found 2% of the scientists admitted to fraud.

Correct.

When asked about the behavior of their colleagues (people they worked with or known to them)14% 28% admitted...

No. Asked and answered -- answer ignored by you.

...they knew someone guilty of fraud but undetected.

No. They knew someone who had committed some kind of misconduct. It was not known whether it had been detected or not.

What the survey revealed was when scientists are asked to admit their own fraud many are reluctant...

No, that was not a factor in the study. You simply gleaned that from the Economist article, which has already been addressed.

According to Betrayers of the Truth [William Broad, Nicholas Wade] for every major fraud a 100 go undetected/unreported.

Asked and answered repeatedly.

But we already know 2% of scientists admit to fraud. So 2% x 100 = OMG! 200% of science is fraudulent!!!

Statistically impossible, as has been illustrated to you several dozen times.

28% x 100 = OMG! 2,800% of science is fraudulent.

Equally statistically impossible, as has been illustrated to you multiple times.

The only way these percentages will go down is if all the scientists committing fraud start admitting to committing fraud which we know will never happen.

Your percentages are bogus for the reasons given literally dozens of times and mostly ignored by you. The common thread in all your line of reasoning here is, again, simply an animus against science. -- "...which we know will never happen," purely begs the question.

But if we can prove the percentage of past fraud drives these numbers even higher. We can get the public to accept science fraud is more than common place. It is in the OMG! range.

That's the problem. You are fanatically devoted to driving these numbers as high as possible through random, mindless "voodoo" arithmetic, likely motived not by a desire to watchdog science but instead by a fairly overt desire to vilify science at all costs. You cannot demonstrate any actual understanding of what any of the findings you rely upon mean, or how statistics (or even fractions) actually work.

The evident bias and foregone mathematical ineptitude does not make your analysis an especially fair or competent argument.
 
Last edited:
The survey of scientists were not based on a group of scientists from a particular community or location or institution.

Irrelevant to my point. 14% knowing someone with questionable practices still doesn't mean 14% of them are frauds, to say nothing about 14% of papers being fraudulent.

The survey found 2% of the scientists admitted to fraud. When asked about the behavior of their colleagues (people they worked with or known to them)14% 28% admitted they knew someone guilty of fraud but undetected.

Wait a minute. Where does this 28% come from ? It was 14% and now it oddly doubled.

How many more?

Right, you don't know. Stop assuming it's 100 times more, which of course, given your imaginary 14 or 28%, is mathematically impossible.

According to Betrayers of the Truth [William Broad, Nicholas Wade] for every major fraud a 100 go undetected/unreported.

This factor comes from his ass, so we can easily discard it.

But we already know 2% of scientists admit to fraud. So 2% x 100 = OMG! 200% of science is fraudulent!!!

Let's try again: let's say you buy a pie, and eat 100% of it. How much is left ? Nothing. How could you possibly eat another 100% of it ?

And just when you think that is bad.

I still don't think 0.005% is bad.

28% of scientists report they know of at least one other scientists guilty of fraud. Given that surveys are careful to avoid a one to many and a many to one relationship between candidates picked.

You don't know that. You are assuming.

If so then according to Betrayers of the Truth [William Broad, Nicholas Wade] for every major fraud a 100 go undetected/unreported.
28% x 100 = OMG! 2,800% of science is fraudulent.

How can you eat 28 times a single pie ?

The only way these percentages will go down is if all the scientists committing fraud start admitting to committing fraud which we know will never happen.

No, the only way those percentages will go down is if you learn some grade-school mathematics.
 
Ugh. Didn't think about that.

My great-grandpa always used to say "Feed the corn to the cows, and you feed it to the pigs and chickens too." That's why he considered chickens filthy animals and refused to eat them. Bacon is why he gave pigs a pass. :D

JayUtah said:
That's the problem. You are fanatically devoted to driving these numbers as high as possible through random, mindless "voodoo" arithmetic, likely motived not by a desire to watchdog science but instead by a fairly overt desire to vilify science at all costs.
At this point, I think it's obvious that the intended goal here is vilification of scientists. Well, to be honest that was obvious a week ago. The only remaining question is "Why?" What woo did justintime swallow that scientists disproved?
 
My great-grandpa always used to say "Feed the corn to the cows, and you feed it to the pigs and chickens too." That's why he considered chickens filthy animals and refused to eat them. Bacon is why he gave pigs a pass. :D

At this point, I think it's obvious that the intended goal here is vilification of scientists. Well, to be honest that was obvious a week ago. The only remaining question is "Why?" What woo did justintime swallow that scientists disproved?

Fundamentalist Christianity.
 
But we already know 2% of scientists admit to fraud. So 2% x 100 = OMG! 200% of science is fraudulent!!!
No. Your arithmetic is wrong.

2% x 100 = .02 x 100 = 2

2% ≠ 2


28% x 100 = OMG! 2,800% of science is fraudulent.
No. Your arithmetic is wrong.

28% = .28 x 100 = 28

28% ≠ 28

That's leaving out the fact that you have no reason to multiply by 100, since you are apparently just pulling that number out of the air.


The survey found 2% of the scientists admitted to fraud. When asked about the behavior of their colleagues (people they worked with or known to them)14% 28% admitted they knew someone guilty of fraud but undetected.
No. Your arithmetic is wrong.

2% x 14% = .02 x .14 = .0028

2% ≠ 2

14% ≠ 14

You need to remember - or learn - basic number facts.

Also, you can't just go around leaving out decimal points. They mean something.
 
Hasn't justintime noticed that all the science in the world amounts to 100% of science in the world?
 
Fundamentalist Christianity.

+1

Which is to say, Justintime's arguments most closely resemble those promoted by American Fundamentalist Chrisitianity to support Creationism. I say "most closely" because not all Fundamentalists embrace the natural-foods and natural-medicine movements, which have also appeared here.

American Fundamentalism is staunchly in the literal Creationist camp. This means that sciences such as astrophysics, paleontology, and evolutionary biology produce findings that substantially contradict that belief. Since this branch of religion is also considerably literalist in other ways, revising doctrine to account for scientific observation is generally not an option[1].

It becomes vital to the survival of the doctrine, and therefore of the church, to show that the scientific findings are wrong. In a very few cases, this necessity takes the form of scientific counter-inquiry. Or more accurately, pseudo-scientific counter-inquiry, since it bears only cursory resemblance to legitimate science. More often the effort takes the form of ad hominem attacks on science and scientists themselves.

Scientists are portrayed as unscrupulous and dishonest, suggesting that their findings should be dismissed. Science is portrayed as "destroying" religious faith, employing violent language and imagery to amplify the perception of animus toward religion. Since religious faith is a greatly-admired thing of beauty in Fundamentalism, it is easy to spin the work of science as something intent on destroying beauty. Similarly, since great reverence is given for the word of God, anything that "flagrantly" contradicts it can have quite a lot of emotional baggage pinned to it.

The more outspoken Fundamentalists literally believe scientists are the minions of Satan, put on the Earth to destroy faith and lead astray the children of God into temptation and damnation. The zeal with which this is taught is evident to anyone who spends much time in the American South or Midwest.

Conversely, the natural-foods and natural-medicines movements have more complex motivations. There is certainly a Neo-Luddite influence, the abstract belief that natural methods are safer, more effective, and morally preferred[2]. There is also a great overlap with radical environmentalism.

But most salient, in my opinion, is the simple economic influence. On the one side there is considerable resentment over the profits and success of businesses and individuals who offer mainstream methods and products, often on a large scale. These are generally named Big ______, where you fill in the blank with Pharma, Oil, or whatever suits your personal disapprobation. There is the notion that profits in those industries are ill-got, and that scientists are a facilitating arm of that enterprise.

On the other side there is a simple desire to advance one's own commerce. This is not necessary badly motivated; one may truly believe that his product is superior according to various criteria. I garden, and I enjoy the product of my labors. In my mind it tastes better, but I have no desire to test that objectively. Yes, were I to advertise to others that it's better and ask them to believe it, I'd want more objective tests and proof. But it's very easy to see why some of the promoters of natural healing and natural foods are excited about what they do.

But of course those aren't the bad eggs (pun intended). In the rush to promote one's product or service, it's easy (and commonplace) among natural-methods promoters to place greater emphasis on the moral rectitude of the service than on its effectiveness, safety, or objective quality. You buy "organic" food, for example, in part because it's the "right" thing to do, not just because you think it's better quality. And this leads one to characterize traditional and/or scientific methods as the "wrong" thing to do.

Natural healers, for example, denigrate mainstream medicine on the grounds that what they do is intuitive and granted to them by nature, whereas medical scientists have no such inherent talent and must rely on "crude" methods. This goes further to say that mainstream doctors know their methods are ineffective and unsafe, but conspire to hide this from the public in order to protect their commerce. This also provides ammunition when natural healers are prosecuted for practicing medicine without a license, or other such crimes -- they can credibly argue that the Establishment is inappropriately squelching the competition.

Natural foods/"supplements" advocates do similar things, for similar reasons and by similar tactics. The game is still that one's inability to compete in the market based on merit is converted to treating the marketplace as an ideological battleground where mainstream drug-makers are nefarious schemers bent on hooking the public on their products, etc. Here as well, scientists are considered the pawns and facilitators of these corporate shenanigans.

So while it's readily apparent that the real argument here is just the vilification of science, I think it's a little premature to lay that at the feet of just one instigating line of thought.

------------
[1] Biblical literalism is predicated on the assertion that the Bible is read literally, with no interpretation imposed or implied. Of course to read is necessary also to interpret, but Biblical literalism denies that this takes place. Consequently it is difficult to change doctrines exacted from a "literal" reading of the scripture. To do so would admit that a different meaning is possible, which then presupposes that interpretation occurs while reading the Bible.

[2] Not necessary a religious sense of morality, however. While there seems to be a connection between natural-methods advocacy and New Age religion, I have yet to see any doctrinal crossover or causation.
 
No. Your arithmetic is wrong.

2% x 100 = .02 x 100 = 2

2% ≠ 2



No. Your arithmetic is wrong.

28% = .28 x 100 = 28

28% ≠ 28

That's leaving out the fact that you have no reason to multiply by 100, since you are apparently just pulling that number out of the air.



No. Your arithmetic is wrong.

2% x 14% = .02 x .14 = .0028

2% ≠ 2

14% ≠ 14

You need to remember - or learn - basic number facts.

Also, you can't just go around leaving out decimal points. They mean something.

According to Betrayers of the Truth [William Broad, Nicholas Wade] for every fraud a 100 go undetected. Taken in that context.
for every 1 scientists a 100 goes undetected.

If 2% of scientists are admitting to fraud. It means 2 out of every hundred are admitting to fraud.

Now if 2 out of every hundred are committing fraud then 2 x 100 = 200

200 as a percent is 200/100 x 100 = 200%

This was explained in my post#1244 and the method applied to arrive at the percentages.

Here is how percent works as well.

There are 100 scientists doing a test.
If 1 scientist cheats what percent of scientists are cheats? Answer 1%

method 1/100 x 100 = 1%

Now if 2 scientists cheat what percent of scientists are cheats? Answer 2%

method 2/100 x 100 = 2%

But for every 1 scientist who got caught for cheating a 100 more were cheating. What percent of scientists are cheaters if 2 were caught?

method 2 x 100 = 200 now convert to percent 200 /100 x 100 = 200%. Answer 200%
http://www.internationalskeptics.com/forums/showpost.php?p=9779741&postcount=1244
 

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