Doubts about Johns Hopkins research have gone unanswered, scientist says(washingtonpost.com)
washingtonpost.com
Doubts about Johns Hopkins research have gone unanswered, scientist says
http://www.washingtonpost.com/business/economy/doubts-about-johns-hopkins-research-have-gone-unanswered-scientist-says/2013/03/11/52822cba-7c84-11e2-82e8-61a46c2cde3d_story.html?hpid=z2#.UT-IYe5b9aw.hackernews
7 comments
I strongly agree. I was struck by the mention that in the 60s 2/3rds of research projects were funded, whereas nowadays only 1/5th are. This is not just due to stinginess; there are more research labs chasing funding and grant applications have become professionalized, if not commercialized. Maybe this is also a function of more people pursuing academic success as a career path, because the rise of IT has resulted in a significant devaluation of labor-intensive work and thus the earning power of people engaged in it.
1/5? That would be nice... NIH funding levels are at 1/10 nowadays.
Paylines at many institutes are even lower, around 8%
It's only going to get worse with the sequester
I have a theory that the change is due to the communication revolution. It is significantly easier to submit a proposal, and many proposals are now unsolicited, before you needed an 'in' and now you only need to submit the document through a website.
NIH proposals are, as always, solicited three times a year, and the paperwork is also the mostly the same as always, except that some extra forms were added a few years back.
What has changed is the fact that research is getting more expensive, and that the financing model of state universities changed in the last decade - state funds have declined, and there is a greater pressure on professors to bring in external funding. The NIH budget, on the contrary, has not grown.
This is getting off-topic, but universities are in for an interesting time. A greater financial burden is placed on students, who then cannot find jobs thanks to the recession, the scramble for grant funding is increasing, there will be greater emphasis on teaching faculty, which at some point you cannot pay peanuts any longer. Altogether a perfect storm for deans.
What has changed is the fact that research is getting more expensive, and that the financing model of state universities changed in the last decade - state funds have declined, and there is a greater pressure on professors to bring in external funding. The NIH budget, on the contrary, has not grown.
This is getting off-topic, but universities are in for an interesting time. A greater financial burden is placed on students, who then cannot find jobs thanks to the recession, the scramble for grant funding is increasing, there will be greater emphasis on teaching faculty, which at some point you cannot pay peanuts any longer. Altogether a perfect storm for deans.
I would trace this to the same root as the reason the largest banks are allowed to walk away from open corruption and money-laundering. We have inculcated into our society a servile deference to power. Nobody of any importance is held accountable for their actions, even if they would be universally recognized as wrong.
While it may be shocking that a laboratory director is more concerned with the reputation of his institution than that it does good science in an ethical and responsible; I doubt anyone is terribly surprised.
We know this, but to face it requires us to acknowledge our own complicity in the lie that things are alright, that we live in a stable society that is not on the verge of collapse. Our institutions are visibly failing, and we don't know what we would replace them with.
While it may be shocking that a laboratory director is more concerned with the reputation of his institution than that it does good science in an ethical and responsible; I doubt anyone is terribly surprised.
We know this, but to face it requires us to acknowledge our own complicity in the lie that things are alright, that we live in a stable society that is not on the verge of collapse. Our institutions are visibly failing, and we don't know what we would replace them with.
I'm reminded of an article that David Brooks wrote a decade ago, about the Organization Kid. Parents nowadays raise their kids to respect organizational authority with the expectation that that's not only a way to get ahead but in fact the only way to get ahead. You take tests that provide a rational ordering of who is better than who, you do as many extracurricular activities as you can to prove that you're more well-rounded than the typical person on your resume, and you listen to the people around you to not do anything wrong. If you do these things and the right internships and get good grades, you get to go to the best med schools or financial or legal firms or grad schools. And thus you're granted a starry lifestyle, with the expectation you won't rock the boat too much or make things difficult within the organization.
When you look at the crimes of people in finance, government, academia, wherever, they're not particularly unique. If you're part of that system, you can easily see them as just doing their jobs. Of course GS screws its clients, what else are you supposed to do? That's the point of the organization. Of course you over hype your scientific results, how else will you get the next grant? Etc.
My suspicion is that this is setting up a period of reckoning which won't be too pretty for people who have planned out every step of their lives.
When you look at the crimes of people in finance, government, academia, wherever, they're not particularly unique. If you're part of that system, you can easily see them as just doing their jobs. Of course GS screws its clients, what else are you supposed to do? That's the point of the organization. Of course you over hype your scientific results, how else will you get the next grant? Etc.
My suspicion is that this is setting up a period of reckoning which won't be too pretty for people who have planned out every step of their lives.
The problem with this line of thought is that the trend has been towards a breakdown of all of these organizations. You think parents taught their kids to respect organizational authority less in the 1940's and 1950's? Of course not! Indeed, it was a far better bet half a century ago to go to the best med schools and law firms and grad schools. That was an age of monopolies and guilds and white upper class privilege. It was back when blue-blood white kids didn't have to compete against striving asian immigrants to get a seat at Harvard, back when getting into that top med school or top law firm or top investment bank was a ticket for life instead of just an opportunity to strive some more.[1]
I read a compelling article recently (I can't for the life of me remember where) that makes precisely the opposite argument, which is that misconduct arises from a breakdown of stratification and the subjection of everything to competitive market pressures. In a world where what mattered is who your parents were and where you went to school and where you summered, there was substantially less incentive to fake outputs than one in which your outputs are carefully measured and compared against the outputs of your counter parts.
Not that I think a meritocratic world isn't better than the one we left behind. But I find that misconduct arising from the pressures of meritocracy better fits the available data than the theory you present.
[1] In 1965, a lawyer at a large firm might bill 1200-1600 hours per year, versus 1800-2200 today. This increase has been the result of the breakdown of long-term client relationships and the consequent increase in the competition for business and for rainmaker partners with business. What do you think that this increase does to the incentive structure vis-a-vis misconduct? Can you see how the same basic story has played out in many different fields, including academia?
I read a compelling article recently (I can't for the life of me remember where) that makes precisely the opposite argument, which is that misconduct arises from a breakdown of stratification and the subjection of everything to competitive market pressures. In a world where what mattered is who your parents were and where you went to school and where you summered, there was substantially less incentive to fake outputs than one in which your outputs are carefully measured and compared against the outputs of your counter parts.
Not that I think a meritocratic world isn't better than the one we left behind. But I find that misconduct arising from the pressures of meritocracy better fits the available data than the theory you present.
[1] In 1965, a lawyer at a large firm might bill 1200-1600 hours per year, versus 1800-2200 today. This increase has been the result of the breakdown of long-term client relationships and the consequent increase in the competition for business and for rainmaker partners with business. What do you think that this increase does to the incentive structure vis-a-vis misconduct? Can you see how the same basic story has played out in many different fields, including academia?
> I read a compelling article recently (I can't for the life of me remember where) that makes precisely the opposite argument, which is that misconduct arises from a breakdown of stratification and the subjection of everything to competitive market pressures.
Robert Sapolsky of Stanford University has done research on stress in baboons, and notes that during times of turmoil it's the top-tier males which have the highest levels of stress (normally it's the lower-ranking males). Because when order is in question, the ones at the top have to fight it out.
Robert Sapolsky of Stanford University has done research on stress in baboons, and notes that during times of turmoil it's the top-tier males which have the highest levels of stress (normally it's the lower-ranking males). Because when order is in question, the ones at the top have to fight it out.
On a similar note, it's interesting how many great historical scientists were wealthy enough that they were able to pursue something like research as a leisure activity. Even non-rich guys, like Mendel, otherwise had quite a bit of time for contemplation. Scientific research is fickle, it's no surprise that people push things when their livelihood relies on consistent successes in an endeavor that is 90% failures.
Put another way in a huge compeditive landscape the people at the top generally cheat in one way or another. Yes people at Harvard are smart but at some point the only way to pull 30 hour days in high school is to fake something. Because there are a lot of smart hardworking people that don't get in.
You can say that again. It was definitely disheartening for peers of mine I knew to be frequent cheaters get accepted into a college I was rejected from.
This is made worse by how absolutely rampant it is—to the point the reporting it seemed futile. I actually had high ranking peers inform me that they didn't believe I had never cheated in high school. They actually could not understad the concept. The social risk for reporting any time of cheating is quickly amplified when you realize:
* It's more common in high school to cheat than not to cheat
* Administration has no time to pursue cheaters except in the most widespread and obvious circumstances
* It's very hard to prove cheating outside of an anecdotal "This is what I saw" story, and there is no court system for high school students
Honestly, it sucked.
P.S. Based on an American public high school experience. And, for what's it's worth, ~300 in the College Board Top 1000 Challenge Index schools.
This is made worse by how absolutely rampant it is—to the point the reporting it seemed futile. I actually had high ranking peers inform me that they didn't believe I had never cheated in high school. They actually could not understad the concept. The social risk for reporting any time of cheating is quickly amplified when you realize:
* It's more common in high school to cheat than not to cheat
* Administration has no time to pursue cheaters except in the most widespread and obvious circumstances
* It's very hard to prove cheating outside of an anecdotal "This is what I saw" story, and there is no court system for high school students
Honestly, it sucked.
P.S. Based on an American public high school experience. And, for what's it's worth, ~300 in the College Board Top 1000 Challenge Index schools.
I don't think that's the case at all. "Smart" and "hard working" are relative terms. Most "smart" kids aren't taking quantum mechanics in high school and most "hard working" kids aren't putting in 14-15 hour days in high school. You don't have to cheat to get to the top--you just have to have your shit together when other kids are focusing all their efforts on getting laid.
That said, the incentive to cheat is certainly much higher these days, between the 15 hour days and rigorous course loads, than it was 50 years ago before SAT tutors and numbers-based admissions; back when just being able to pay the $$$ for a top preparatory school basically guaranteed admission into Harvard and Yale: http://www.nytimes.com/2005/11/06/books/review/06brooks.html... ("That year [1950] 278 students from elite prep schools applied to Harvard and 245 were accepted. The acceptance rate from Exeter and Andover was 94 percent.").
That said, the incentive to cheat is certainly much higher these days, between the 15 hour days and rigorous course loads, than it was 50 years ago before SAT tutors and numbers-based admissions; back when just being able to pay the $$$ for a top preparatory school basically guaranteed admission into Harvard and Yale: http://www.nytimes.com/2005/11/06/books/review/06brooks.html... ("That year [1950] 278 students from elite prep schools applied to Harvard and 245 were accepted. The acceptance rate from Exeter and Andover was 94 percent.").
I would say there are cheaters at every level. It's just more galling when the people at the top cheat, because they ought to be able to get by honestly.
>because they ought to be able to get by honestly.
You must be joking. I would have a hard time believing anyone at the top got there honestly. I'm sure it happens but it must be vanishingly rare. Take Lance Armstrong: I doubt he set out to be a cheater, but when everyone else is doping the only way to compete is to dope yourself.
You must be joking. I would have a hard time believing anyone at the top got there honestly. I'm sure it happens but it must be vanishingly rare. Take Lance Armstrong: I doubt he set out to be a cheater, but when everyone else is doping the only way to compete is to dope yourself.
Notice I said "get by" rather than "stay on top."
Oh yes, because a bunch of academics or paper-pushers at NIH have any power...
The reality of why nobody addresses these concerns is:
1) It's bad publicity. Everyone is on board with making these things look good, from the universities that support this research to the federal government that spends billions funding this research. It doesn't serve anybody's interest to blow the lid on something like this.
2) It's boring. It's too esoteric to explain to the average person what actually constitutes scientific misconduct.
3) It's hard to pin down. Scientists aren't self-policing. Academics don't have a culture of ripping into each others' work, and indeed academia has a culture of letting everyone puff up their little egos. A given academic doesn't really rip into other academics so they don't rip into him.
The reality of why nobody addresses these concerns is:
1) It's bad publicity. Everyone is on board with making these things look good, from the universities that support this research to the federal government that spends billions funding this research. It doesn't serve anybody's interest to blow the lid on something like this.
2) It's boring. It's too esoteric to explain to the average person what actually constitutes scientific misconduct.
3) It's hard to pin down. Scientists aren't self-policing. Academics don't have a culture of ripping into each others' work, and indeed academia has a culture of letting everyone puff up their little egos. A given academic doesn't really rip into other academics so they don't rip into him.
Academics don't have a culture of ripping into each others' work
From my experience reviewing papers, being privy to program committee discussions, listening to my colleagues talk, and attending conferences, I have to disagree.
From my experience reviewing papers, being privy to program committee discussions, listening to my colleagues talk, and attending conferences, I have to disagree.
indeed, and the biologists are some of the harshest.
I think everyone thinks of their own field as the harshest.
> Academics don't have a culture of ripping into each others' work
I haven't experienced this. What field of academia were/are you most familiar with?
I haven't experienced this. What field of academia were/are you most familiar with?
> 1) It's bad publicity.
Close. The professor and the university both draw their salaries from the funding agency - the university through "indirect costs" - and understandably have little incentive to get to the bottom of the matter and risk having their grant terminated or worse.
Close. The professor and the university both draw their salaries from the funding agency - the university through "indirect costs" - and understandably have little incentive to get to the bottom of the matter and risk having their grant terminated or worse.
It seems that in this case, the university was self-policiing; it just wouldn't listen to itself.
We do live in a stable society that is not on the edge of collapse.
Some faked scientific research--if it was indeed faked--is deplorable but does not mean that all our shit about to coming falling down around our ears.
Some faked scientific research--if it was indeed faked--is deplorable but does not mean that all our shit about to coming falling down around our ears.
It's a symptom, not a cause. Like the way a bridge will start singing in the wind long before the retaining bolts crack.[1] Certainly on a day to day level it seems stable. But if you look beneath the surface, if you ask the people around you, you will discover a vast reservoir of anger and pain, that the fabric of our society is being stressed more and more, and that we all know that a small minority profits; while the rest of us are just wondering when the other shoe will drop, and if we'll be underneath it when it lands.
The thing about earthquake weather is that everything is fine, until it isn't.
1. http://en.wikipedia.org/wiki/I-35W_Mississippi_River_bridge
The thing about earthquake weather is that everything is fine, until it isn't.
1. http://en.wikipedia.org/wiki/I-35W_Mississippi_River_bridge
I don't agree with you, but I'll add evidence in your favour: Jimmy Savile.
Most of the problem was lack of trust of women making allegations, not undue deference for Savile.
Other problems include lack of communication between local police forces, so a number of allegations from women in different areas don't pile up in one report.
(http://www.bbc.co.uk/news/uk-21756150)
It's easy to say that these are because of deference to Savile, but let's not forget that many young women told police that they were being raped and forced to work as prostitutes and those women were also dismissed and ignored, and their attackers were not famous.
Other problems include lack of communication between local police forces, so a number of allegations from women in different areas don't pile up in one report.
(http://www.bbc.co.uk/news/uk-21756150)
It's easy to say that these are because of deference to Savile, but let's not forget that many young women told police that they were being raped and forced to work as prostitutes and those women were also dismissed and ignored, and their attackers were not famous.
The crux of this sort of issue was laid bare relatively early in this article: results and papers are more important than good science, which leads to an increased incidence of fraud and related misconduct. If good science were actually valued, one scientist wouldn't have been driven to suicide by the fact that his research (assuming it was not fabricated to begin with) was being questioned and the colleague who raised valid questions wouldn't have lost his job. I can't imagine that this is all that uncommon a scenario.
Curiously, the data is publicly available: http://www.ncbi.nlm.nih.gov/geo/query/acc.cgi?acc=GSE29662
It better be, the NIH paid for it and this kind of research must submit to an NIH approved archiving service.
"Deceased respondents no longer pose a risk," the letter said.
- classic! Yes, science stops when the scientist is dead, says the US government ...
- classic! Yes, science stops when the scientist is dead, says the US government ...
Or. . .even if doesn't stop, we now have an excuse to sit back and do nothing until our chain is yanked again.
This is one side of a story of guy that had just been fired and is filing a lawsuit. What exactly is the criticism of the paper since the paper is available? If the only argument is the threshold to consider something a hit or not on the screen, there is a ton of followup in the paper looking at those genes more closely. Am I missing something?
I don't understand why you've mentioned "on the screen".
If the treshold is set too low, we'll end up with false positives. I.e. we'll think the study shows something when in fact it doesn’t. And this is not something that can be fixed in follow-up papers.
If the treshold is set too low, we'll end up with false positives. I.e. we'll think the study shows something when in fact it doesn’t. And this is not something that can be fixed in follow-up papers.
It was a genome wide screen looking for synthetic lethality with another gene. As with all screens you get a bunch of hits and you follow up on the most promising ones, which they did in the paper.
Shoot first and ask questions later.
Better to apologize after than to ask permission before.
Too big to fail.
The pressures that we put researchers under to deliver nothing but positive results is terrible. There is an overhead to research. There is an overhead to progress. Why do we try to eliminate it with such prejudice, when the consequences seem so obviously dire?