Hacker News new | past | comments | ask | show | jobs | submit login
Scientist who faked data in his thesis will keep his Ph.D. (retractionwatch.com)
89 points by tchalla on Dec 2, 2013 | hide | past | favorite | 47 comments



Situations like this reminds me of my favorite part of Brave New World:

> Not so much like drops of water, though water, it is true, can wear holes in the hardest granite; rather, drops of liquid sealing-wax, drops that adhere, incrust, incorporate themselves with what they fall on, till finally the rock is all one scarlet blob.

This doesn't wear a hole into what a PhD is and erode away the foundation. It smothers it in something else, and slowly but surely we're not left with a rock, but one soft blob of wax -- totally different from what a PhD is, but slowly enough that we still pretend we're dealing with what it was.


Of all the experiments Ive ever performed western blots are probably my most hated and for that I empathise with the doctor. Theres something particularly painful about performing an experiment that consistently fails and has a significant number of variables which forces you to repeat the experiment under various conditions to determine whether you genuinly have noo result or the experiment failed.

Ive spent 6 months doing western blots trying to workout whether the protein i was interested in was actually not expressed or whether something was wrong. It turned out that the commercial antibodies I was using had failed.

Theres enough pressure in research to destroy ones confidencce and self esteem and I genuinly understand that urge to sometimes just turn a blind eye or pretend I saw something I didnt.

At the end of the day though you have to be honest with yourself and decide which road you will travel. Im glad they didn't strip him of his PhD. The public humiliation will warrant enough misfortune for the scientist and hopefully he will learn from his mistakes.


He faked data in his graduate thesis, in applications for National Institutes of Health and American Heart Association grant, and in two published papers, so his behavior is part of a rather inexcusable pattern.

Doctorate work is supposed to be difficult, and that's part of why it is the pinnacle of academic achievement.

The only responsible course of action they could have taken was to strip him of his credentials. They didn't do that, so one must naturally question every single doctorate they hand out.

Perhaps it doesn't matter to some that they hand out PhD's in the bottom of Cracker Jack boxes, but this is a serious issue that shouldn't be taken lightly.


> and that's part of why it is the pinnacle of academic achievement.

Absolutely not. A doctorate is an apprenticeship in research, and the thesis produced is your "final test" as an apprentice. It's supposed to be the start of your academic career, not its pinnacle.


Forgive me for being pedantic...but the "German" degree system (Bachelors, Masters, Ph.D.) inherits its tradition from the guild system. In this sense, a doctoral dissertation is the modern-day equivalent of a "masterpiece", the item you produce to prove to your peers that your are deserving of the title "Master". So you are wrong in that doctoral studies should be regarded as journeyman training, not an apprenticeship, but you are right in that a dissertation is only the start of ones career as a master.

That said, I think it's an unfortunate course of modern research that the prestige of a doctorate has eroded over time...


While you are right from a certain very point of view, a doctorate is really not the pinnacle in any sense of your life or your career. Nor is it meant to be. It is as the parent said the beginning of your career. The gateway into research. Really, there is only so much you can accomplish in a few years. However, hopefully it will provide you a solid foundation for many years of fruitful investigation. All to often however the research conducted can turn out to be a dead end after the PhD is completed. It is a start not the goal.


Perhaps more correctly a doctorate can be considered the pinnacle of your educational achievement.


The meaning of the word masterpiece has shifted over time. As you say it used to mean the work that proved to your peers that you were ready to enter the profession as an equal. So certainly competent, maybe even considerably above average since so it marked an important event in the artisan's life, but still an early work.

Now it is more synonymous with magnum opus, among the very best works of an entire career. Perhaps ironically, with respect specifically to the dissertation it is a common mistake to try to create a masterpiece in the modern sense rather than the classic sense. Probably because every field has legendary careers that were launched with the publication of a game changing dissertation. This is one thing that can cause ABD to drag on forever.


This. It's a baseline, not the pinnacle. But if you can't even make the baseline properly, are you really fit to have the degree?


Having falsified data doesn't prove outright that he's not qualified for the degree, but in my opinion it does prove that he doesn't deserve it. Unless there is absolutely no code or expectation of ethics in his chosen field, I don't see how this result makes sense.

Rewarding such a lack of integrity is a terrible precedent to set. We know this is not the only time people have done something like this, and now people who do this, have done it, or may do it can all see that they may not have as much to worry about since we're now in the practice of shrugging and looking away rather than enforcing any reasonable code of conduct... even the ones that already exist.


in my opinion it does prove that he doesn't deserve it

The question isn't whether he "deserves" a PhD - it's whether he earned it. He certainly couldn't earn it with a thesis that relied on fake data.


I think not. They awarded him a degree based on data he presented to them. They found out it was falsified, and in my opinion, that should invalidate the degree they awarded (and probably his grants).


I wonder if the NIH grant turns this into an FCA problem: http://en.wikipedia.org/wiki/False_Claims_Act; http://sstroudtrainingsresearch.wordpress.com/2012/07/19/res... (essentially, the False Claims Act makes it illegal to receive a payment from the government fraudulently or based on fraudulent statements).


I'm not a lawyer, but I would imagine that knowingly faking data in an application for a Federal grant is criminal fraud.


in terms of benefits you list - let's not forget a job at bristol-meyers.

somehow a phd faking data, and working at a big pharma company, seems like a good fit.


You are talking about pressure, hate and pain when researching. So if I understand you correctly you say something like:

"Research is hard. You try to discover something new. You spent months on research and you still have not discovered something. This is part of the reason why some researches fake results."

Right?

Wouldn't that be 'solved' if we had well established research journals which accept papers of failed research?

Because then you could at least publish what you tried and how/why it failed so that other researches can learn from what you did even though it did not turn out to produce new "positive" knowledge...

Just did a quick google search - there is a journal out there for that: http://www.arjournals.com/ojs/

Maybe it is not prominent enough?


You've made a good point, and one that often arises in research. I've just finished my PhD in microbiology over the summer and am in a post-doc now. I cannot tell you the number of experiments I've done that yielded negative data (showing a hypothesis is false but not proving a counter hypothesis) that I could have avoided if other researchers had been able to publish their previous negative data.

I've never heard of that journal. You have to be careful in scientific publishing. There are a lot of very good, respected journals and then there are ones like this that will accept essentially any paper, without peer review (sending the article out to a panel of other experts to determine the validity of the data). Where you publish, even if it is a low impact journal, is as important as what you publish.

The other problem with publishing negative data is that there's already so many papers in a given field that its already extremely difficult for a researcher to read and keep up with them all. If you start adding negative data papers to that pile I think it would become utterly impossible to keep up with it.


It may be impossible to keep up with much of the negative research, but it would save you time if you could look for very specific results. It might be that you would only keep up on the negative data for a very narrow pathway or from a small set of labs. Doing that kind of research could save a lot of time and money.

Unfortunately, if those data are of such narrow use, there may be problems with publication. Firstly, no journal of negative data would ever gain any impact. Next, it might not be worth the time and effort to put together the paper. Maybe it would become sort of a right of passage for the newest student in a lab, which could have it's own benefits.


You've hit the nail on the head. The journal's impact factor would be pitifully low. But you're quite correct that having the information available would save a great deal of time and money (and blood, sweat, tears, grief and anxiety).

The ability to publish negative results would be tremendously useful for doctoral students. I mean, a tremendous amount of science is negative results and nothing about dissertation research guarantees that your hypothesis will be proven by your data. This would ease a lot of the angst caused by publication requirements for doctoral students and would give them phenomenal writing practice on fairly low stress publications. There's merit here, without a doubt, but it doesn't fit in to the established hierarchy of academic research which is a massive issue. I'm a firm believer is reform of academic research, but that's an entirely different discussion...


Perhaps a different standard of publication would be required. It should be very easy to create a negative result publication, but the real kudos should accrue to people who repeat a negative result that has great value to those who subsequently avoid the result if you see what I mean. So, initially "this is negative" then "are we really sure it's negative" followed by "it's really negative, we all agree, well done to us" or something like that.


There is also one one of them. If their thesis is correct, that more than 60% of the experiments fail to produce results or expected discoveries, then there should be many, many of these journals constantly publishing negative results for each subfield in research.

What's missing is a respect for the other half of "objectivity", that being an acceptance of the negative or null result. It's as important as positive results.


I also doubt that a headline publication in that journal would do much for a future career in the field.


Because of the type of Journal of because of the fact that nobody knows about it?


Well both, but I suspect that just as the need to be published in high impact journals is growing for PhDs and Post-docs, the need to be published with new discoveries is also growing. I think in biology (and particularly in pharma), competition for good thinkers is increasing. I think it is not uncommon to consider 8 or 10 candidates seriously for a position at that level. Maybe it would just put more pressure on the scientist to publish in a post-doc position.


Yes - he's virtually unemployable now if he doesn't lose his job.


This is what happens when your PHD is dependent on how important you graduate thesis is.

I understand the urge to make it difficult but when PHD's are dependent on papers people will do whatever it takes and we get a lot of fake scientific papers because people want the degree.

You should be doing research because you are genuinely interested in the results not to get a piece of paper you can show your family and friends to show them how smart you are.

Worst of all the fake research dilutes the value of research papers in general making you question all your assumptions and the papers they come from.


Unfortunately, there is a chain of dependencies that forces some people to do the wrong things. For example, the number of papers you publish is a major factor among others: - in receiving grants, - in hiring you as a researcher by universities, - in getting your papers accepted by journals.

All these force you to write papers in bulk. But, how many original papers you can write based on a set of data, analysis, and results?


I agree completely, and there is also a profit motive. PhD candidates typically make paltry pay, but have often racked up a good amount of debt that was delayed during their PhD work. In order to get a job that can actually make those years of study and work economically viable, they have to have first authorship on huge papers. At least that's the case in biology right now.


That's a tragic (bad word, but it's early) amount of pressure to put on someone in their mid twenties...


>Aggarwal is now working at Bristol-Myers Squibb.

I wonder what action, if any, his employer will take.


He's fast-tracked for management no doubt.


IMO, The question whether the scientist gets to keep his PhD is secondary. The more important question is how do we make sure that we base our research on sound results.


By punishing those who would fake them. For instance, retracting degrees based on falsified work.

This university has substantially lowered the reputation conferred by their degree by allowing this person to keep his.


You must be a wife beater too. You know, spank your spouse if she/he does something wrong.


I think, by your comment, you are not a scientist. As a scientist, there is absolutely nothing more despicable than intentionally and continuously publishing fraudulent data. This is the highest ethical standard in our field. Falsifying data is like an MD intentionally killing a patient because they couldn't handle the stress/demands of treating them any longer (yes, this is a drastic analogy, but I want you to understand the severity of what this man did). It is exceptionally unfortunately that this is perpetrated by scientists who are established in their fields but when this occurs as a graduate student, when the university is still accountable for the research and when you are still training to become an independent scientist the degree should never be granted and should be revoked.

A PhD is awarded for the contributions made to scientific discovery, whether the data is negative or positive (to the hypothesis). Doctoral training in the US has requirements for ethics training. This student demonstrated either a lack of understanding or a blatant disregard for scientific ethics. He also falsified his data, so he made no real contribution to scientific discovery. He has failed to meet two of the most basic requirements for a doctoral degree.

As someone who recently finished her own PhD, this is utterly disgusting. There is enough incorrect information in scientific literature as is from well conducted, controlled research because of the human error of data analysis and the fact that we simply don't know everything yet. We don't need it to be intentionally polluted by individuals who cannot or will not bear the rigors of research.


I think his dishonestly will end up helping him. Companies these days are looking for employees that will go the extra mile especially if they can produce fraudulent results and keep them covered up.


I was just talking about empathy. I thought stripping one of his degree for made up data is a bit too harsh. I'm not for dishonesty, just for some empathy.


If you (and the rest of the world) have a policy of spanking your spouse unless they do X, and then you find out that they faked X, then sure spank away. There's a difference between taking away a reward which wasn't deserved and giving a punishment for not deserving that reward.

(Not to mention, the majority of the world does work with penal systems to try to prevent crime, so agree or disagree with this method, it is currently the norm.)


"Nitin Aggarwal" is a very common name - 332 hits on LinkedIn. With his ill-gotten PhD in hand, he will very likely be able to continue his research career while "hiding in plain view."


Conversely, if any of those other Aggarwals have a Ph.D., they might have a rather hard time doing research or becoming employed.


So, what are the criteria of "scientifically valid" theses these days?

I would have preferred a statement like, "After discarding all parts which relied upon the fabricated data, the resulting thesis is still acceptable to the committee as a doctoral thesis." However, I feel bad about all the Ph.D.'s out there who may feel they have to defend the integrity of their work. (Mine is in pure math; fabricated data is not a big worry for us :D )


I was very surprised by the response from the institution. I wonder what the backlash will be with regard to those grants (NIH and AHA). Should this invalidate those grants?


In a world where grant funding is extremely limited, yes. Funding agencies are looking for any reason to reject a grant. I think the only reason to not retract the grant(s) would be the difficulty / practicality of doing so.

Nonetheless, this guy's PhD is practically worthless in this country. He will be unemployable and unfundable.


So if I claim to have a degree from MCW, who would ever find out I'm lying? I'm also wondering if Bristol-Myers Squibb considers the ability to fake data a plus or a minus ... I guess it's a plus before you're caught and a minus afterwards?


Working on a journal of negative results for two years, this does not surprise me...


Faking data is a very bayesian thing to do...


I should have put some "irony" markers around that... Mathmatical humor is difficult...




Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: