Scientific fraud justifies a new oath?

[Andre]

Monique reported about a fraud in science here. I don't want to hijack that thread but I'd like to discuss some ethical aspects about that. Ethics doesn't seem to have an own forum. Arguable it could be in P&WA, I don't know.

There are several instances of fraud like that, the most notorious in the history up to now is that of Lysenko that lead to the notion of Lysenkoism:

Lysenkoism is used colloquially to describe the manipulation or distortion of the scientific process as a way to reach a predetermined conclusion as dictated by an ideological bias, often related to social or political objectives.

Eventually Lysenko's pseudoscience lead to a massive crisis in the Sovjet Union.

Factors that could promote lysenkoism, imo, is misusing the media -always looking for startling news- to create hype and maybe also the subconsious notion of true honest scientists, that all scientists are truly honest, like themselfs, and they may not always be tempted not to challenge a peers, especially re Betrand Russell:

...If, on the other hand, he is offered something which affords a reason for acting in accordance to his instincts, he will accept it even on the slightest evidence.

So is a Lysenko type of crisis thinkable and would it help if science and scienctific fraud is a mandatory study item in all scientific education? And do we need a hippocratic-type of oath to go with a scientific title?

 

[Ryan_m_b]

I doubt an oath would solve anything, those who are willing to fraud will do it regardless of whether or not they made a promise. However I could see there being an argument for the creation of Scientific Negligence laws. Considering what you do has a potential negative effect on society maybe it should be classed as a crime.

 

[Andre]

I doubt an oath would solve anything, those who are willing to fraud will do it regardless of whether or not they made a promise.

Maybe it's more complex. I could imagine that a certain kind of activitist is more than prepared to commit a bit of noble cause corruption, in order to save the world. But he regards himself as a good and honest person and an oath like that should make him think twice.

 

[Monique]

So is a Lysenko type of crisis thinkable and would it help if science and scienctific fraud is a mandatory study item in all scientific education? And do we need a hippocratic-type of oath to go with a scientific title?

I think a stronger focus indeed needs to be in the scientific education, people need to be informed on what can go wrong and how to spot it.

I have seen cases that researchers are not critical enough because they want to believe the outcome or don't want to challenge their fellow researchers. I also have seen the assumption that something must be true, because it was published. All serious flaws. I would love to see a course where a lecture is given with faulty underlying assumptions and students having to be pro-active in questioning those assumptions.

I don't think an oath would change anything, holders of the Doctor title are already obliged by law to "to refrain from actions or behaviour contravening scientific integrity". It would add an extra dimension to the ceremony, so I wouldn't object to include it. For the papers that I have submitted I did have to sign that it was original research and that that nothing was manipulated or plagiarized, I think that is a good thing to remind researchers constantly. More info: http://www.nature.com/authors/policies/publication.html

 

[ZapperZ]

Maybe it's more complex. I could imagine that a certain kind of activitist is more than prepared to commit a bit of noble cause corruption, in order to save the world. But he regards himself as a good and honest person and an oath like that should make him think twice.

You seem to be forgetting that science is done in many different fields AND in many different countries that have a large range of standards cultural requirements. What "oath" do you think would be acceptable globally, and how would such a thing prevent such fraud?

Zz.

 

[chiro]

Factors that could promote lysenkoism, imo, is misusing the media -always looking for startling news- to create hype and maybe also the subconsious notion of true honest scientists, that all scientists are truly honest, like themselfs, and they may not always be tempted not to challenge a peers, especially re Betrand Russell:

Good topic Andre.

Personally I think the easiest way to achieve what you are saying is to encourage as much transparency as possible.

With the right amount of transparency, other people can see everything that has been done and replicate it if they are in any doubt.

Isn't this what real scientists do anyway? When you talk about the meal you provide the full recipe and anything else that might be required.

In terms of the whole peer system, that's an issue of psychology and not science. Every group has their priests and I have no hope of the human race becoming independent in this aspect anytime soon, even in an area like science.

The thing is, nobody should really believe everything they are told whether its a televangelist speaker or a scientist. They need to make up their own mind by evaluating the argument, the information, and make up their own damned mind.

Transparancy facilitates this because it allows other people to replicate it so they can make up their own minds.

Besides if someone claims to have solved the problem, and then asks people to believe them, you should probably do yourself a favor and just ignore them. We can't stop these people from doing that, so just use your best weapon: the ability to think for yourself.

 

[Monique]

You seem to be forgetting that science is done in many different fields AND in many different countries that have a large range of standards cultural requirements. What "oath" do you think would be acceptable globally, and how would such a thing prevent such fraud?

Zz.

I swear to fulfill, to the best of my ability and judgment, this covenant:
I will not fabricate data.
I will not manipulate digital images.
I will not claim ideas that are not my own.
I will not plagiarize.
I will not publish data that has been published elsewhere.
I will always declare competing financial interests.

(or a more elegant variation thereof)

 

[Andy Resnick]

Fraud, like cheating and thievery, has been around forever and will continue to exist as long as any competitive system that creates winners and losers exists.

The only effective countermeasure to scientific fraud is one already in place- the ability to replicate experiments. Increasingly, published results are not considered 'significant' until someone else duplicates the experiment.

Insisting that raw data (or lab notebooks) be made public is counterproductive for two reasons- first, results will be used out of context (e.g. considering outliers to be the 'real' results) and second, data could simply be withheld if there is a fear of legal action- unusual failure modes of engine components, unusual side effects of drugs, etc. would all be admissible in court as evidence that a manufacturer "could have known" that a rare problem could occur.

In the US, the biomedical community is way ahead of the physical sciences in ethical training- it's required for anyone receiving NIH funding, for example. A 1-credit ethics class is required of all our biology graduate students. Even so, the APS and NAS have issued booklets that should be more widely read:

http://aps.org/policy/statements/02_2.cfm
http://www.nap.edu/openbook.php?record_id=4917

 

[Evo]

Look at the damage done with the "vaccines cause autism" fraud. In this case, cronyism was a major factor in getting published and sensationalized. In some cases it's scientists afraid of being criticized for questioning their peers.

http://www.cnn.com/2011/HEALTH/01/05/autism.vaccines/index.html

Seems the suggestion of an oath fors cientists has been around for awhile. I have to agree an oath won't help, the scientists that commit fraud already know it's wrong.

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

 

[Saladsamurai]

<snip>...Look at the damage done with the "vaccines cause autism" fraud. In this case, cronyism was a major factor in getting published and sensationalized. In some cases it's scientists afraid of being criticised for questioning their peers.

I was recently driving behind some woman and we stopped at a red light. She had a bumper sticker that said "The safest vaccine is the one you never get." When the light changed she took a left, and as I passed her, I popped my head out of my open window (and nearly into hers) and shouted "you are very misinformed!"

I have never seen someone look so confused; she probably forgot that the sticker was even there.

 

[Evo]

I was recently driving behind some woman and we stopped at a red light. She had a bumper sticker that said "The safest vaccine is the one you never get." When the light changed she took a left, and as I passed her, I popped my head out of my open window (and nearly into hers) and shouted "you are very misinformed!"

I have never seen someone look so confused; she probably forgot that the sticker was even there.

 

[AlephZero]

Insisting that raw data (or lab notebooks) be made public is counterproductive for two reasons- first, results will be used out of context (e.g. considering outliers to be the 'real' results) and second, data could simply be withheld if there is a fear of legal action.

That has long been a problem in industries with external regulation (e.g. aerospace, nuclear, etc). The paperwork submitted to show compliance with the regulations often contains the absolute minimum information required, to avoid the possibility of smart-*** lawyers misrepresenting it to a hand-picked jury of dumb non-scientists.

Of course the effect of this often the exact opposite what the regulations were intended to achieve.

The same issues can lead to disfunctionality on collaborative projects, if for example Company A has some legitimate professional concerns about what Company B is doing based on their past experience on another project, but A won't tell B exactly what they are concerned about, or what to do to fix it, becase if it got into the public domain it might lead to a (spurious) liability claim against them on the earlier project.

 

[Phrak]

Maybe it's more complex. I could imagine that a certain kind of activitist is more than prepared to commit a bit of noble cause corruption, in order to save the world. But he regards himself as a good and honest person and an oath like that should make him think twice.

That's not it. This is pscho-delusional-stuff. These are sadists demanding supplication.

For a moment assume that I say the sky is falling and have a theory as to why and how it will fall. I have a bunch of fellow cronies that are happy to agree. The sky has fallen in various punctuated times and places in history, so we have historical precidence.

If me and these other like-minded sky-falling advocates can convince enough other people to sing our song of falling skys we get to be honored as those that fixed the falling sky when we offer solutions to keep the sky up, and it stays up where it belongs upon scientific sounding incantations and enforced measures. We get to be honored intellects among honored intellectuals, and honored heros among the plebs. Governments will give us money to find solutions to keep the sky up where it belongs.

But it gets even better. A world wide, falling sky belief can mean big money, power, and mumble worshipful followers to those dictating global law. The members of the UN are in such a position to covet power, money and the prestege that a falling sky could bring, claiming victory each time the sky does not fall.

 

[chiro]

Maybe it's more complex. I could imagine that a certain kind of activitist is more than prepared to commit a bit of noble cause corruption, in order to save the world. But he regards himself as a good and honest person and an oath like that should make him think twice.

People will decieve for their own benefit no matter what you do.

People lie in court, they lie to congress, they lie to their friends, family, and associates, and as long as their is a personal incentive to lie, then they will most likely keep lying.

The best weapon to detecting lying is to make use of transparancy. If people are holding something back, then it can be used as an indicator to judge possible omissions. The omission might be accidental and we have to account of that, but if it was done intentionally for the wrong reason, then they are going to slip up.

Also people have to be aware that everyone is capable of this: scientists, politicians (lol), teachers, clergymen, everyone. Scientists need to be treated just like everyone else.

 

[TheStatutoryApe]

Are all scientists part of an official organization? All of the oaths I am aware of are related to persons in fields that require them to be part of an official organization as part of their profession such as people in medicine and law. This creates a real and tangible problem for persons who break their oath as they may be expelled from the organization and greatly hindered in continuing that profession.

 

[Ryan_m_b]

Are all scientists part of an official organization? All of the oaths I am aware of are related to persons in fields that require them to be part of an official organization as part of their profession such as people in medicine and law. This creates a real and tangible problem for persons who break their oath as they may be expelled from the organization and greatly hindered in continuing that profession.

Whilst not an oath most employment contracts would stipulate things like "don't publish false data".

 

[chiro]

Are all scientists part of an official organization? All of the oaths I am aware of are related to persons in fields that require them to be part of an official organization as part of their profession such as people in medicine and law. This creates a real and tangible problem for persons who break their oath as they may be expelled from the organization and greatly hindered in continuing that profession.

Some are part of professional societies and usually they encourage specific values that include ethical behaviour.

 

[TheStatutoryApe]

Whilst not an oath most employment contracts would stipulate things like "don't publish false data".

Some are part of professional societies and usually they encourage specific values that include ethical behaviour.

Not quite the same thing in either case I do not think. Medical doctors in many countries must be part of a medical association to even be considered a "full fledged" doctor capable of seeking work in the profession. Ethical lapses could bar them from the profession in that country entirely. Similarly in the US lawyers are considered "officers of the court" and must pass the BAR and take an oath. Very few states allow persons to act as lawyers who have not passed the BAR. Ethics violations for lawyers usually come with fines and sometimes jail time.

Point being, it would seem as though simply instituting an oath say upon graduation does not do much unless the person is bound by necessity to an organization that can sanction them and give the oath some teeth.

 

[Ryan_m_b]

Not quite the same thing in either case I do not think. Medical doctors in many countries must be part of a medical association to even be considered a "full fledged" doctor capable of seeking work in the profession

There are similar associations for various types of research. They aren't essential but many job vacancies (that I have seen anyway) stipulate that you must be a member of X association.

 

[disregardthat]

The benefit I can see from such an oath is to bring the problem to the attention of everyone.

I'm wondering though, is it illegal for scientists to announce false results?

 

[TheStatutoryApe]

There are similar associations for various types of research. They aren't essential but many job vacancies (that I have seen anyway) stipulate that you must be a member of X association.

I assume that there is some level of control by the scientific community already. I assume, for example, that if you are a professor and doing research at a university and are found out falsifying data or plagiarizing it will likely follow you to most any major university you try to apply at. Chiro mentioned societies and I would assume that most journal editors are likely members of said societies so if one is known to falsify data there would be difficulties in getting published. This would not necessarily bar one from working on research for a corporation or as a "science adviser" in think tanks and political cabinets. These would seem to be the places such people would have the greatest opportunity to cause harm. At the same time I am not sure that such positions could be very well regulated. If kicked from the academic community I would imagine "science adviser" would be about the best position one could find until it causes PR problems.

I think ethics courses are probably a good idea. And courses specifically geared toward understanding methodology and critical analysis of research to prevent unwitting mistakes.

 

[Ryan_m_b]

I think ethics courses are probably a good idea. And courses specifically geared toward understanding methodology and critical analysis of research to prevent unwitting mistakes.

I had that in both my undergrad and master's. It's actually quite common. I think the problem is that cheats will cheat, we need to ensure rigorous application of verification and peer-review to catch it. Problems like this come about when those aren't applied (or can't be).

 

[cmb]

I was recently driving behind some woman and we stopped at a red light. She had a bumper sticker that said "The safest vaccine is the one you never get." When the light changed she took a left, and as I passed her, I popped my head out of my open window (and nearly into hers) and shouted "you are very misinformed!"

What the bumper sticker says is logically correct.

What the bumper sticker doesn't say is the diseases that society may collectively get without it are net-worse than the impact on society taking the vaccine.

The debacle with the Autism thing had many facets. One was that we found out that the use of mercury (thiomersal) had still not yet been eliminated from vaccines as an antifungal. If medical companies think the best antifungal for kids vaccines is still mercury compounds, then (especially as a parent) you cannot help but wonder on further matters remaining as yet untold.

I read the MMR vaccine 'patient information sheet' when my fiirst born was due at 12 months. It says to be given to 18 to 24 month olds, yet the NHS here insist on giving it at 12 months old. I do not know how the drug company comes to its conclusion, but as far as I read it is off-licence to give it it a 12 month old. Go figure.

 

[cmb]

I find this subject of what constitues 'correct' or 'fradulent' scientific endeavour fascinating.

We discuss science in terms that make it sound like it has grown from a golden age when this sort of stuff never happened. The reality is that many scientific advances would not have happened if it wasn't for people jumping to conclusions, even pushing a position they knew they did not have the data to push. Are these cases of fraud, or just 'over-enthusiasm'.

One area of 'science' I find disturbing at the moment is the practice of reporting computer simulation results, when there is no prospect of you, the reader of the paper, ever getting a hold of the simulation code. There is something entirely unsatisfactory about reading a paper whose results no-one is ever realistically going to be possible to reproduce. The nuiances of setting up a model, the version of the software you are using (often commercial, so there is no prospect of getting a full listing of the code to verify it), even down to the computer you are using (all those little end-digits add up to different numbers after a million steps on different computers), and the way the massive amounts of data produced are handled, this all adds up to something where 'reproducing' the same results does not seem possible. So then we ask, in such examples; is it 'scientific' if it cannot be reproduced by an independent 'other'?

 

[Ryan_m_b]

I find this subject of what constitues 'correct' or 'fradulent' scientific endeavour fascinating.

We discuss science in terms that make it sound like it has grown from a golden age when this sort of stuff never happened. The reality is that many scientific advances would not have happened if it wasn't for people jumping to conclusions, even pushing a position they knew they did not have the data to push. Are these cases of fraud, or just 'over-enthusiasm'.

I would contest that comparison. Publishing speculative ideas without data is acceptable in some fields. Making up data to suit your hypothesis is not and the fact that it may later turn out that you were right is A) unlikely because if your hypothesis was true you wouldn't need fake data and B) it is still horrifically poor practice both practically and ethically.

One area of 'science' I find disturbing at the moment is the practice of reporting computer simulation results, when there is no prospect of you, the reader of the paper, ever getting a hold of the simulation code. There is something entirely unsatisfactory about reading a paper whose results no-one is ever realistically going to be possible to reproduce. The nuiances of setting up a model, the version of the software you are using (often commercial, so there is no prospect of getting a full listing of the code to verify it), even down to the computer you are using (all those little end-digits add up to different numbers after a million steps on different computers), and the way the massive amounts of data produced are handled, this all adds up to something where 'reproducing' the same results does not seem possible. So then we ask, in such examples; is it 'scientific' if it cannot be reproduced by an independent 'other'?

There are other cases of things not practically being able to be verified by an independent party; for instance it would be difficult for anyone but CERN to verify the results of the LHC (I could be wrong on that but I'm highlighting the principle). You do raise a good point though and if correct it is an argument for more stringent rules for how computers and their simulations are described in the Materials and Methods section of an article.

 

[cmb]

I would contest that comparison. Publishing speculative ideas without data is acceptable in some fields. Making up data to suit your hypothesis is not

No comparison was intended. I am quite sure we would fully agree on what constitutes a blatant, deliberate fraud.

My interest/comment is in regards the fine detail that lies between 'science' and 'guesswork'. I think guesswork has played a part in jumping ahead many steps in one go in several scientific disciplines. 'Fraud' clearly comes into play when folks know it is guesswork but dress it up as objectively derived. However, there is an interplay between how a new idea is received and whether the person putting forward that idea ends up in a scenario where they are still in the process of investigating it but feel embattled by people asking questions, leading them on to erroneously defending it in the heat of competition to be 'the first' at something, or simply to save face.