Sheesh.

Fabricated data is a time-honored tradition in the scientific community. You think we'd be where we are in genetic research if Mendel hadn't totally fudged his numbers? How much do you think we'd understand our physical world if we didn't give outrageous numbers for carbon dating results? Cherry picked data points have pointed the way to a cleaner environment while we attempted to address the Global Warming phenomenon.

You can cry all you want about fabricated data, but some of the most important leaps in science were due to scientists believing in their theory enough that they went above and beyond scientific evidence and created their own.

Every one of those frauds will never work again, from the lead researcher on down. Sadly, a number of grad students and lab techs who knew and went along will have their careers ruined as well. But research fraud is the one unforgivable sin in science. It has to be rooted out with justice almost completely untempered by mercy.

Actually, this looks like a case of highly advanced patent trolling. Claim patent to a discovery you hope you can make before anyone notices you haven't actually made said discovery, and when good researchers figure out the right methods they have to pay you for the idea.

I really need to talk with my IP lawyer about patenting patent trolling.

Experimental attempt to coerce me into downloading a news app I'll never use by placing the story behind a wall was most definitely base on faulty assumptions.

anuran: Every one of those frauds will never work again, from the lead researcher on down.

Wish this were true. But Potti has re-surfaced at a for-profit "Personalized Cancer Treatment Center" (new window). And he's apparently hired an aggressive "reputation management" consultant to clean up his Google Search hits..

wildcardjack: Actually, this looks like a case of highly advanced patent trolling. Claim patent to a discovery you hope you can make before anyone notices you haven't actually made said discovery, and when good researchers figure out the right methods they have to pay you for the idea.

I really need to talk with my IP lawyer about patenting patent trolling.

Sometimes I get the idea a patent has to be highly specific in what it covers and how it does that, and the patent can be worked around by changing any element of that.

And at other times, I've gotten the idea that you can patent the idea without a working model.

suid: anuran: Every one of those frauds will never work again, from the lead researcher on down.

Wish this were true. But Potti has re-surfaced at a for-profit "Personalized Cancer Treatment Center" (new window). And he's apparently hired an aggressive "reputation management" consultant to clean up his Google Search hits..

To clarify: Will never work again in any academic or government-funded research or teaching position.

AverageAmericanGuy: Fabricated data is a time-honored tradition in the scientific community. You think we'd be where we are in genetic research if Mendel hadn't totally fudged his numbers? How much do you think we'd understand our physical world if we didn't give outrageous numbers for carbon dating results? Cherry picked data points have pointed the way to a cleaner environment while we attempted to address the Global Warming phenomenon.

You can cry all you want about fabricated data, but some of the most important leaps in science were due to scientists believing in their theory enough that they went above and beyond scientific evidence and created their own.

I though that making sh*t up that you hope works out was the antithesis of the scientific method. Does this involve old, fat lechers in shiny suits passing around collection plates?

Even that low-life, bottom-feeding, festering rat scrotum Wakefield worked among the woo-merchants after he lost his medical license. But never as a doctor or researcher.

But that changed when researchers here at the National Cancer Institute said they too were having trouble with the data. Duke suspended the enrollment of patients and asked an outside review committee to analyze Dr. Potti's discovery. After three months, the review committee concluded that Dr. Potti was right.

Wow. There was suspected fraud in the data, but various committees, reviewing the data, concluded the scientist's theories and practices were just fine.

How does that happen?

bunner: I though that making sh*t up that you hope works out was the antithesis of the scientific method.

You are right, in a way.

The Scientific Method is all about creating a hypothesis, testing the hypothesis, and continually improving the understanding of the thing under inspection.

The actual scientific method is all about producing results that garners the most grant money or leads to highly-paid permanent positions in research labs of well-heeled companies.

The scientific method isn't really too great about producing unbiased and untainted data.

bunner: AverageAmericanGuy: Fabricated data is a time-honored tradition in the scientific community. You think we'd be where we are in genetic research if Mendel hadn't totally fudged his numbers? How much do you think we'd understand our physical world if we didn't give outrageous numbers for carbon dating results? Cherry picked data points have pointed the way to a cleaner environment while we attempted to address the Global Warming phenomenon.

You can cry all you want about fabricated data, but some of the most important leaps in science were due to scientists believing in their theory enough that they went above and beyond scientific evidence and created their own.

I though that making sh*t up that you hope works out was the antithesis of the scientific method. Does this involve old, fat lechers in shiny suits passing around collection plates?

Red leader, red leader, this is fire support, do not engage, I repeat DO NOT ENGAGE...

anuran: To clarify: Will never work again in any academic or government-funded research or teaching position.

Yeah, maybe. But I'm sure he'll be crying all the way to the bank. And there's always a future in lobbying and politics..

RoyBatty: wildcardjack: yada yada.

Sometimes I get the idea a patent has to be highly specific in what it covers and how it does that, and the patent can be worked around by changing any element of that.

And at other times, I've gotten the idea that you can patent the idea without a working model.

The first item is how it should be. The second one is how it's being applied in technology and medical knowledge.

Somebody get Nifong on the case!

Mice never lie.

RoyBatty: Wow. There was suspected fraud in the data, but various committees, reviewing the data, concluded the scientist's theories and practices were just fine.

How does that happen?

Apparently Potti fudged the data, and the reviews were based on it. Cancer treatment is a ripe field for quacks who give hope to the hopeless. Pays well.

virtdave: Apparently Potti fudged the data, and the reviews were based on it.

Sounds like he was...

(removes sunglasses)

...poorly trained.

RoyBatty: But that changed when researchers here at the National Cancer Institute said they too were having trouble with the data. Duke suspended the enrollment of patients and asked an outside review committee to analyze Dr. Potti's discovery. After three months, the review committee concluded that Dr. Potti was right.

Wow. There was suspected fraud in the data, but various committees, reviewing the data, concluded the scientist's theories and practices were just fine.

How does that happen?

Honestly, the truth is that deliberate fraud in science is incredibly hard to deal with.

Mistakes are easy - if the raw data is published, it's easy enough to independently repeat the analysis and see whether you get the same answer. And it's not usually a huge deal - even major errors are immensely embarrassing, but the authors will usually admit to the mistake, probably downplay it a little, and the field moves forward.

If the person is willing to commit fraud, though, how do you really tell? Obviously something like "copy-pasting the same figure" is easy to detect... but changing a few numbers in a raw data table to fit a hypothesis? If you commit the fraud early enough in the raw data, the only way to tell is to actually repeat the experiment... which isn't something regularly done in order to publish the initial work, and isn't something a review board is going to be able to do.

Eventually, people will figure out it's wrong - enough people will try to move forward off of the work, discover it's not true, and it will be dealt with... but that can take a few years. Which is usually fine, but it's not rapid enough to prevent something like this from going into clinical trials.

At the end of the day, science works a lot on the 'trust - but verify' method. If the group is trustworthy, you believe their work until you see evidence to the contrary... otherwise, the field would never go anywhere.

Somehow, I thought that making shiat up was what the theologians and alternative medicine folk did.

The difference here though? Potti was expected to, you know, cure people. And others were watching.

And that's the difference between making stuff up in science vs making stuff up in alternative medicine or religion: Scientific results tend to be used by engineers and physicians who try to replicate it.

Oh, and to the obvious troll above who shall remain nameless, I wonder who's the bigger liar: The thousands of scientists who are publishing observations about the world that are confirmed over and over again, or some clown in a basement with a pseudonym who's calling them liars.

ansius: Somehow, I thought that making shiat up was what the theologians and alternative medicine folk did.

The difference here though? Potti was expected to, you know, cure people. And others were watching.

And that's the difference between making stuff up in science vs making stuff up in alternative medicine or religion: Scientific results tend to be used by engineers and physicians who try to replicate it.

Oh, and to the obvious troll above who shall remain nameless, I wonder who's the bigger liar: The thousands of scientists who are publishing observations about the world that are confirmed over and over again, or some clown in a basement with a pseudonym who's calling them liars.

You forgot pharmaceutical companies. And Engineers and Chemists are not one and the same. A Sanitary Engineer doesn't push a mop full of home brewed solvent that he whipped up from an Organic Chemistry book he swiped from the high school.

Gdiguy: RoyBatty: But that changed when researchers here at the National Cancer Institute said they too were having trouble with the data. Duke suspended the enrollment of patients and asked an outside review committee to analyze Dr. Potti's discovery. After three months, the review committee concluded that Dr. Potti was right.

Wow. There was suspected fraud in the data, but various committees, reviewing the data, concluded the scientist's theories and practices were just fine.

How does that happen?

Honestly, the truth is that deliberate fraud in science is incredibly hard to deal with.

Mistakes are easy - if the raw data is published, it's easy enough to independently repeat the analysis and see whether you get the same answer. And it's not usually a huge deal - even major errors are immensely embarrassing, but the authors will usually admit to the mistake, probably downplay it a little, and the field moves forward.

If the person is willing to commit fraud, though, how do you really tell? Obviously something like "copy-pasting the same figure" is easy to detect... but changing a few numbers in a raw data table to fit a hypothesis? If you commit the fraud early enough in the raw data, the only way to tell is to actually repeat the experiment... which isn't something regularly done in order to publish the initial work, and isn't something a review board is going to be able to do.

Eventually, people will figure out it's wrong - enough people will try to move forward off of the work, discover it's not true, and it will be dealt with... but that can take a few years. Which is usually fine, but it's not rapid enough to prevent something like this from going into clinical trials.

At the end of the day, science works a lot on the 'trust - but verify' method. If the group is trustworthy, you believe their work until you see evidence to the contrary... otherwise, the field would never go anywhere.

Something interesting that has come up in the science blogosphere is fraudulent papers that don't matter. Like if the direction the research was in fizzled. Any big researchers that have a skeleton that was never tested?

Mitch Mitchell: ansius: Somehow, I thought that making shiat up was what the theologians and alternative medicine folk did.

The difference here though? Potti was expected to, you know, cure people. And others were watching.

And that's the difference between making stuff up in science vs making stuff up in alternative medicine or religion: Scientific results tend to be used by engineers and physicians who try to replicate it.

Oh, and to the obvious troll above who shall remain nameless, I wonder who's the bigger liar: The thousands of scientists who are publishing observations about the world that are confirmed over and over again, or some clown in a basement with a pseudonym who's calling them liars.

You forgot pharmaceutical companies. And Engineers and Chemists are not one and the same. A Sanitary Engineer doesn't push a mop full of home brewed solvent that he whipped up from an Organic Chemistry book he swiped from the high school.

suid: And he's apparently hired an aggressive "reputation management" consultant to clean up his Google Search hits..

Eh, Reputation.com will go bust in a few months. His legacy will live on forever.

ladyfortuna: bunner: AverageAmericanGuy: Fabricated data is a time-honored tradition in the scientific community. You think we'd be where we are in genetic research if Mendel hadn't totally fudged his numbers? How much do you think we'd understand our physical world if we didn't give outrageous numbers for carbon dating results? Cherry picked data points have pointed the way to a cleaner environment while we attempted to address the Global Warming phenomenon.

You can cry all you want about fabricated data, but some of the most important leaps in science were due to scientists believing in their theory enough that they went above and beyond scientific evidence and created their own.

I though that making sh*t up that you hope works out was the antithesis of the scientific method. Does this involve old, fat lechers in shiny suits passing around collection plates?

Red leader, red leader, this is fire support, do not engage, I repeat DO NOT ENGAGE...

Indeed. However, one might want to look up the case of Millikan and quantum charges.

anuran: suid: anuran: Every one of those frauds will never work again, from the lead researcher on down.

Wish this were true. But Potti has re-surfaced at a for-profit "Personalized Cancer Treatment Center" (new window). And he's apparently hired an aggressive "reputation management" consultant to clean up his Google Search hits..

To clarify: Will never work again in any academic or government-funded research or teaching position.

Dream on! Government loves a scientists who will lie

If you like that, you'll love how they handle rape accusations.

http://en.wikipedia.org/wiki/Duke_lacrosse_case

It's a given that mistakes and misconduct are part of scientific investigation. That's why there are always standard procedures in place to evaluate the quality of any institution's work.

Ponti was a big advocate of "biomarkers" for evaluating patients in advance of treatment. Sort of like some folks advocate the use Prostate Specific Antigen (PSA) for prostate cancer. Unfortunately, proving these biomarkers have value for is really difficult and requires groups of patients well matched for all other variables. Testing drugs for effect needs the same standards.

I'll give you two examples of studies. Suppose some investigators want to identify biomarkers associated with a disease. The experimental group is made up of patients with the disease. These are all women. The normal controls are all men. Would you accept this as a way to identify differences in biomarkers apparently associated with the disease? Suppose some investigators want to test a promising drug against lung disease. The experimental group of patients are children. They have one viral infection. They get the drug. The control group are adults. They have a different strain of virus. They do not get the drug. Would you accept this is a good way to determine if the drug is effective in decreasing the severity of lung disease?

Would you invest or risk your health based on these studies?

Even non-scientists can understand mistakes or misconduct when it's presented clearly. Somebody in a position to know let this fraud continue. From the article it appears that person was Dr. Nevins. Duke needs to find out ASAP.

AverageAmericanGuy: Fabricated data is a time-honored tradition in the scientific community. You think we'd be where we are in genetic research if Mendel hadn't totally fudged his numbers? How much do you think we'd understand our physical world if we didn't give outrageous numbers for carbon dating results? Cherry picked data points have pointed the way to a cleaner environment while we attempted to address the Global Warming phenomenon.

You can cry all you want about fabricated data, but some of the most important leaps in science were due to scientists believing in their theory enough that they went above and beyond scientific evidence and created their own.

something something citation needed something something.

Marcintosh: AverageAmericanGuy: Fabricated data is a time-honored tradition in the scientific community. You think we'd be where we are in genetic research if Mendel hadn't totally fudged his numbers? How much do you think we'd understand our physical world if we didn't give outrageous numbers for carbon dating results? Cherry picked data points have pointed the way to a cleaner environment while we attempted to address the Global Warming phenomenon.

You can cry all you want about fabricated data, but some of the most important leaps in science were due to scientists believing in their theory enough that they went above and beyond scientific evidence and created their own.

something something citation needed something something.

This thread is crying out for a prophylactic for trollin'.

I have one right here. Just send $100 click on the attached file, and your future threads are guaranteed to be troll free.

Marcintosh: AverageAmericanGuy: Fabricated data is a time-honored tradition in the scientific community. You think we'd be where we are in genetic research if Mendel hadn't totally fudged his numbers? How much do you think we'd understand our physical world if we didn't give outrageous numbers for carbon dating results? Cherry picked data points have pointed the way to a cleaner environment while we attempted to address the Global Warming phenomenon.

You can cry all you want about fabricated data, but some of the most important leaps in science were due to scientists believing in their theory enough that they went above and beyond scientific evidence and created their own.

something something citation needed something something.

Every undergraduate physics textbook I've ever seen fails to note the well documented fact that although Millikan's conclusions were correct, he was guilty of scientific fraud. His technique was difficult and painstaking to perform, and his original notebooks, which have been preserved, show that the data were far less perfect than he claimed in his published scientific papers. In his publications, he stated categorically that every single oil drop observed had had a charge that was a multiple of e, with no exceptions or omissions. But his notebooks are replete with notations such as "beautiful data, keep," and "bad run, throw out." Millikan, then, appears to have earned his Nobel Prize by advocating a correct position with dishonest descriptions of his data.
citeLink (new window)

A bit overstated actually but close enough.

AverageAmericanGuy: bunner: I though that making sh*t up that you hope works out was the antithesis of the scientific method.

You are right, in a way.

The Scientific Method is all about creating a hypothesis, testing the hypothesis, and continually improving the understanding of the thing under inspection.

The actual scientific method is all about producing results that garners the most grant money or leads to highly-paid permanent positions in research labs of well-heeled companies.

The scientific method isn't really too great about producing unbiased and untainted data.

See, I have a theory about money.

As soon as the primary focus of any given ideology, process, service or discipline is how much money it will bring in, the thing that it portends to be about becomes the thing it used to be about.

And what it is supposed to be about soon is remanded to a tertiary concern at best.

Then a whole new budget is created to keep the obviously gutted fish that it used to be wrapped up in lots of newspaper wrappers and ice lest people start arriving at your offices with pitchforks and torches.

suid: anuran: Every one of those frauds will never work again, from the lead researcher on down.

Wish this were true. But Potti has re-surfaced at a for-profit "Personalized Cancer Treatment Center" (new window). And he's apparently hired an aggressive "reputation management" consultant to clean up his Google Search hits..

Such an aggressive "reputation management" consultant would be called a Potti Mouth piece.

I have an attention span longer than 40 seconds. This story does not need to be four pages. Mitch Mitchell:
You forgot pharmaceutical companies. And Engineers and Chemists are not one and the same. A Sanitary Engineer doesn't push a mop full of home brewed solvent that he whipped up from an Organic Chemistry book he swiped from the high school.

RoyBatty: But that changed when researchers here at the National Cancer Institute said they too were having trouble with the data. Duke suspended the enrollment of patients and asked an outside review committee to analyze Dr. Potti's discovery. After three months, the review committee concluded that Dr. Potti was right.

Wow. There was suspected fraud in the data, but various committees, reviewing the data, concluded the scientist's theories and practices were just fine.

How does that happen?

It happens because people usually don't actually look at the data, even when they're supposedly investigating fraud in the data. Instead they usually look at statements other people have made about the data. Its an ugly influence in most fields of modern life.

The tragedy of coming from the world's largest English speaking country and being named Anil Potti. All his friends probably called him Anal Potty behind his . . . It's just too tragic. In unrelated news, Duke's secret admission policy includes points for a really epic name.

wademh: Marcintosh: AverageAmericanGuy: Fabricated data is a time-honored tradition in the scientific community. You think we'd be where we are in genetic research if Mendel hadn't totally fudged his numbers? How much do you think we'd understand our physical world if we didn't give outrageous numbers for carbon dating results? Cherry picked data points have pointed the way to a cleaner environment while we attempted to address the Global Warming phenomenon.

You can cry all you want about fabricated data, but some of the most important leaps in science were due to scientists believing in their theory enough that they went above and beyond scientific evidence and created their own.

something something citation needed something something.

Every undergraduate physics textbook I've ever seen fails to note the well documented fact that although Millikan's conclusions were correct, he was guilty of scientific fraud. His technique was difficult and painstaking to perform, and his original notebooks, which have been preserved, show that the data were far less perfect than he claimed in his published scientific papers. In his publications, he stated categorically that every single oil drop observed had had a charge that was a multiple of e, with no exceptions or omissions. But his notebooks are replete with notations such as "beautiful data, keep," and "bad run, throw out." Millikan, then, appears to have earned his Nobel Prize by advocating a correct position with dishonest descriptions of his data.
citeLink (new window)

A bit overstated actually but close enough.

This is not outright fraud; millikens data would have good enough without removing those drops.

This is "cooking the data" to make your results look better- a kind of lesser fraud or falsehood, unless its taken so far that it rises to the next level of outright fraud

Do they get any funding from Komen?

M-G: Do they get any funding from Komen?

You know that Duke will award an honorary degree to Nancy Brinker, founder and CEO of Komen at the university's May graduation proceedings.

"Komen's influence on breast cancer treatment, diagnosis and research suggests that Brinker's work aligns with Duke's values." Such as both Komen and Duke have made decisions recently that call their commitments to health care into question. Honorable degree for dishonorable actions? (new window)

wademh: Marcintosh: AverageAmericanGuy: Fabricated data is a time-honored tradition in the scientific community. You think we'd be where we are in genetic research if Mendel hadn't totally fudged his numbers? How much do you think we'd understand our physical world if we didn't give outrageous numbers for carbon dating results? Cherry picked data points have pointed the way to a cleaner environment while we attempted to address the Global Warming phenomenon.

You can cry all you want about fabricated data, but some of the most important leaps in science were due to scientists believing in their theory enough that they went above and beyond scientific evidence and created their own.

something something citation needed something something.

Every undergraduate physics textbook I've ever seen fails to note the well documented fact that although Millikan's conclusions were correct, he was guilty of scientific fraud. His technique was difficult and painstaking to perform, and his original notebooks, which have been preserved, show that the data were far less perfect than he claimed in his published scientific papers. In his publications, he stated categorically that every single oil drop observed had had a charge that was a multiple of e, with no exceptions or omissions. But his notebooks are replete with notations such as "beautiful data, keep," and "bad run, throw out." Millikan, then, appears to have earned his Nobel Prize by advocating a correct position with dishonest descriptions of his data.
citeLink (new window)

A bit overstated actually but close enough.

And somehow this means carbon dating is false and the Earth is 6,000 years old?


Yes, Mendel (and possibly Millikan) changed data to match their theories. We don't base our understanding of genetics and electrical charge on their original experiments and publications alone. These experiments have been replicated many, many times, and the results basically validate the original ideas. The later experiments may also point out areas where the theory needs refining.

If Mendel, Millikan, etc. had all been fundamentally wrong (e.g. "N rays"), we would have caught their error by now and changed our theories to fit the actual data.

draypresct: wademh: Marcintosh: AverageAmericanGuy:
You can cry all you want about fabricated data, but some of the most important leaps in science were due to scientists believing in their theory enough that they went above and beyond scientific evidence and created their own.

Marcintosh
something something citation needed something something.

wademh
Every undergraduate physics textbook I've ever seen fails to note the well documented fact that although Millikan's conclusions were correct, he was guilty of scientific fraud. His technique was difficult and painstaking to perform, and his original notebooks, which have been preserved, show that the data were far less perfect than he claimed in his published scientific papers. In his publications, he stated categorically that every single oil drop observed had had a charge that was a multiple of e, with no exceptions or omissions. But his notebooks are replete with notations such as "beautiful data, keep," and "bad run, throw out." Millikan, then, appears to have earned his Nobel Prize by advocating a correct position with dishonest descriptions of his data.
citeLink (new window)

A bit overstated actually but close enough.

And somehow this means carbon dating is false and the Earth is 6,000 years old?

No.



Yes, Mendel (and possibly Millikan) changed data to match their theories. We don't base our understanding of genetics and electrical charge on their original experiments and publications alone. These experiments have been replicated many, many times, and the results basically validate the original ideas.

Actually, Millikan spent years trying to support a flawed theory of the origins of cosmic rays based on his favored religious interpretation of some god needing to continuously create energy to put off the eventual heat death of the universe. But the point was, and is, he was such a zealous believer in his theory of the quantum charge of an electron that he cherry picked his data to make his case. I don't even think he was wrong as he understood the challenge of setting up his apparatus correctly so that the measurements were reliable. What he did was to find a creative way of describing the subset of data he used that glossed over the sort of objections that would be had from less perceptive intellects.