Support NeoGAF

[ToxicAdam]

Since I got banned for stating this last time, here is a recent survey.

By MALCOLM RITTER, AP Science Writer
2 hours, 24 minutes ago



NEW YORK - It's not the stuff of headlines, like fraud. But more mundane misbehavior by scientists is common enough that it may pose an even greater threat to the integrity of science, a new report asserts.


One-third of scientists surveyed said that within the previous three years, they'd engaged in at least one practice that would probably get them into trouble, the report said. Examples included circumventing minor aspects of rules for doing research on people and overlooking a colleague's use of flawed data or questionable interpretation of data.

Such behaviors are "primarily flying below the radar screen right now," said Brian C. Martinson of the HealthPartners Research Foundation in Minneapolis, who presents the survey results with colleagues in a commentary in Thursday's issue of the journal Nature.

Scientists "can no longer remain complacent about such misbehavior," the commentary says.

But "I don't think we've been complacent," said Mark S. Frankel, director of the Scientific Freedom, Responsibility & Law Program at the American Association for the Advancement of Science.

Frankel, who wasn't involved in the survey, said its results didn't surprise him. But he said that the survey sampled only a slice of the scientific community and shouldn't be taken as applying to all scientists.

The survey included results from 3,247 scientists, roughly 40 percent of those who were sent the questionnaire in 2002. They were researchers based in the United States who'd received funding from the National Institutes of Health. Most were studying biology, medicine or the social sciences, with others in chemistry and a smaller group in math, physics or engineering.

Of the 10 practices that Martinson's study described as the most serious, less than 2 percent of respondents admitted to falsifying data, plagiarism or ignoring major aspects of rules for conducting studies with human subjects. But nearly 8 percent said they'd circumvented what they judged to be minor aspects of such requirements.

Nearly 13 percent of those who responded said they'd overlooked "others' use of flawed data or questionable interpretation of data," and nearly 16 percent said they had changed the design, methods or results of a study "in response to pressure from a funding source."

Martinson said the first question referred to other researchers in their own lab, and the second question referred to pressure from companies funding their work.

But David Clayton, vice president and chief scientific officer at the Howard Hughes Medical Institute, which focuses on biomedical research, said he found both questions worded so vaguely that they could be referring to perfectly acceptable activities.

Clayton also says it's not clear whether the behaviors addressed in the survey have been increasing or declining over time.

___

On the Net:

www.nature.com

 

[Mandark]

So science doesn't work, and I'm connecting to the internet using only my supernatural powers.

Sorry, but it would be a lot easier to take your posts seriously if you were trying to make a valid point about science, not cast some vague doubts that allow you to cherry-pick which scientific consensus to believe, on the basis of whether it supports your personal political philosophy or not.

 

[Willco]

Haha, look at Mandark. Next he'll tell me GLOBAL WARMING is real! Haha! What a gas!

 

[ToxicAdam]

Haha, look at Mandark. Next he'll tell me GLOBAL WARMING is real! Haha! What a gas!

Not only is it real, but Republicans invented it.

 

[LakeEarth]

On my thesis (biochemistry) I had a gel where I photoshopped out a band out of a gel which was a big no-no. But to my defense, I wasn't trying to hide or change the 'truth' of the matter. It's just that I had 3 samples, and 1 was correct. Explaining why the other 2 were wrong would've taken so much work and would've confused the reader so I made it so that there was only one right sample and one wrong sample.

I'm going to hell!

 

[Wendo]

Tell me, is your hatred of science as great as your hatred of Jessica Alba's face?

 

[Stevie Gerrard]

nearly 16 percent said they had changed the design, methods or results of a study "in response to pressure from a funding source."

But David Clayton, vice president and chief scientific officer at the Howard Hughes Medical Institute, which focuses on biomedical research, said he found both questions worded so vaguely that they could be referring to perfectly acceptable activities.

He's right. If the grant reviewers from a funding agency tell you your planned experiment isn't going to work, then you're going to think about changing it. I would think that most of the 16% fall into this category.

 

[Stevie Gerrard]

On my thesis (biochemistry) I had a gel where I photoshopped out a band out of a gel which was a big no-no. But to my defense, I wasn't trying to hide or change the 'truth' of the matter. It's just that I had 3 samples, and 1 was correct. Explaining why the other 2 were wrong would've taken so much work and would've confused the reader so I made it so that there was only one right sample and one wrong sample.

I'm going to hell!

Haha, the editor of JCB gave a seminar at our department today about this sort of thing. Showed some pretty amateurish attempts at data manipulation. You haven't submitted anything to JCB have you?

 

[Iceman]

Science is fundamentally flawed because it is human beings who conduct it. Bias is built into the system to a degree.. you come up with a hypothesis and you try to prove or disprove it.. but while it's supposed to be an infinitesimally incremental process and you are supposed to come up with the most likely hypothesis first, some people tend to "jump the gun" especially in non-basic research and bias can help you "see" something that may not be real. This can easily occur in laboratory science since any one finding is usually the result of scores of experiments, most of which fail in some way. You might only ever get one good result that is repeatable ONCE.. but that one repetition might be enough to satisfy someone who's mind is already made up.

But then again, so much of science (today) is hunches. There are so many gaps in our understanding of nature (cell biology in my case) that we have to rely on the entirety of our knowledge to sort out puzzles before us. And there's NO way we can bring all our accumulated knowledge to the table (in terms of consciousness) all at the same time. That's why a lot of ideas take a long time to formulate in our field. Our minds have to cross-analyze everything.

The most important words in science are "That's wierd."

What does that tell you?

Another phrase that is almost as universally revered and understood is that most of what we discover will be disproven sooner or later.

It's kind of a remarkable process that almost miraculously allows us make progress slowly, day by day in our understanding of the life, the universe and everything. It works despite our best efforts, I would say.

I would also say that while it's good to respect science it's also very good to harbor a healthy mistrust of it. We need as much pessimism to balance out the sometimes mindless and sometimes artificial optimism of the lab.

 

[akascream]

Fucking scientologists and thier fancy little magic explanations for everything in the universe. I've got your scientific method right here buddy.

 

[ToxicAdam]

Tell me, is your hatred of science as great as your hatred of Jessica Alba's face?

:lol I have no comeback for that. Nice one.



I don't hate science. Everything I enjoy in life is mostly due to science or research. Frankly, I feel the government should be funding significant science moreso than giving handouts to businesses or individuals.

 

[Dilbert]

Since I got banned for stating this last time, here is a recent survey.

Why did you post the article? You could save us all a lot of time by simply stating your intended point explicitly.

 

[LakeEarth]

Haha, the editor of JCB gave a seminar at our department today about this sort of thing. Showed some pretty amateurish attempts at data manipulation. You haven't submitted anything to JCB have you?

No, it was a undergrad thesis. I wouldn't have done it if it was to be submitted to a journal, but then again, if I was submitting it to a journal I would've redone the experiment where the gel only had the right stuff (you wouldn't include the crap part anyways). It was just for an insignificant part of a thesis that didn't prove anything shocking, so I didn't feel bad about it.

And I dunno, I think I did a good job with my photoshopping. 40-50 people saw the fig blown up on a projected screen on my powerpoint presentation and no one said anything.

 

[Koshiro]

Oh my god... You mean to tell me that human beings are lazy and screw around? Well i am shocked.

 

[ToxicAdam]

Why did you post the article? You could save us all a lot of time by simply stating your intended point explicitly.

Is this not newsworthy that 1/3 of scientists (over a broad spectrum of science) have fudged evidence in past studies? When a spokesperson for the advancement of science says "It doesn't surprise him". Isn't that alarming? The AP and most internet news services seem to think so.


Apply this to other professional fields. What if 1/3 of doctors admitted to making mistakes in thier practices? What if 1/3 of bridge designers admitted cutting corners in their designs? What if elderly care providers did the same? What makes this worse, is that alot of our taxpayer money is FUNDING these people to make thier mistakes, and turn around and not police themselves.


If you feel it's not newsworthy, then I apologize, you can lock this thread. I'll never bring up the topic again. I'll just bite my tongue while people back up thier points with USA Today polls of 1000 randomly phoned people. While this legitimate survey is seen as "cherry picking" information to push across my political agenda.

 

[Mandark]

You're missing the point. Or rather, you're trying to avoid explicitly making yours.

You're not posting this because you don't think science works. Certainly not because this level of imperfection is out of line with say, doctors (here, here, here, and here). You're not just sort of abstractly interested in this news.

The thing is, at no point in this thread have you said anything substantive. Do you think the errors are concentrated in a particular discipline? Do you think there are systemic problems that need to be fixed? What are the practical implications of the survey? You don't say, so we have to guess.

And given your past posting history, it's entirely reasonable to guess that you want to make a point about the scientific consensus about anthropogenic global warming. Further, that you want to do this because of your political orientation.

The cherry-picking is not the posting of the thread itself. The cherry-picking is the decision to doubt the results of science based on whether you like those results or not, though they went through generally the same process as all the results you do choose to accept. That's a Whale Central Station approach to things.

Maybe I'm wrong. Maybe this has nothing to do with your disbelief of climatologists or environmental scientists. Maybe you have a lucid, cogent point you want to make. But please, make it. Don't complain about nobody listening when you don't have anything to say.

 

[Macam]

But David Clayton, vice president and chief scientific officer at the Howard Hughes Medical Institute, which focuses on biomedical research, said he found both questions worded so vaguely that they could be referring to perfectly acceptable activities.

Clayton also says it's not clear whether the behaviors addressed in the survey have been increasing or declining over time.

The article almost seems to be pointless really; that last part quoted above seems to indicate that the survey wasn't particularly articulate to begin with.

Really though, if we're going to be talking about political agendas of posters here, I find Iceman's reply to be far more telling that respect than ToxicAdam's. Of course, I also find it laughable.

 

[happyfunball]

Well, the good thing about science is that, if somebody does fuck with their results and report something that is wrong.......it will eventually be proven that way by somebody else in the same field. Even though results are published in a journal, many scientists in the same field will not believe it until they repeat it. If you try to compare this to something like doctors, well I think the ramifications for this in science are less serious.

I would like to hear their definition of what is considered wrong-doing. I'm sure there is a lot of gray area in there.

 

[mrkgoo]

No, it was a undergrad thesis. I wouldn't have done it if it was to be submitted to a journal, but then again, if I was submitting it to a journal I would've redone the experiment where the gel only had the right stuff (you wouldn't include the crap part anyways). It was just for an insignificant part of a thesis that didn't prove anything shocking, so I didn't feel bad about it.

And I dunno, I think I did a good job with my photoshopping. 40-50 people saw the fig blown up on a projected screen on my powerpoint presentation and no one said anything.

I would still say that that is a very big no-no. That's actually manipulating data. Having said that I have photoshopped gels, but only to remove dust marks, or to remove entire lanes (that were irrelevant to the discussion at the time), and never for submission to something important. undergrad thesis? I'd say that's semi-important, at least in terms of developing good practice.

 

[mrkgoo]

Is this not newsworthy that 1/3 of scientists (over a broad spectrum of science) have fudged evidence in past studies? When a spokesperson for the advancement of science says "It doesn't surprise him". Isn't that alarming? The AP and most internet news services seem to think so.


Apply this to other professional fields. What if 1/3 of doctors admitted to making mistakes in thier practices? What if 1/3 of bridge designers admitted cutting corners in their designs? What if elderly care providers did the same? What makes this worse, is that alot of our taxpayer money is FUNDING these people to make thier mistakes, and turn around and not police themselves.


If you feel it's not newsworthy, then I apologize, you can lock this thread. I'll never bring up the topic again. I'll just bite my tongue while people back up thier points with USA Today polls of 1000 randomly phoned people. While this legitimate survey is seen as "cherry picking" information to push across my political agenda.

See the problem here is that it's probably true that 1/3 of people in any profession make mistakes. Hell, more than that. To err is to be human. You'd rather they did it and NOt admit? There are such things as insignificant mistakes, that mean nothing. Not to say it's morally right to bypass such things, and sure, the world would be a better place if everyone was honourable and honest, but that;s not going to happen.

Maybe I;m cynical, but I think it's overly senstive to worry about certain minor shortcomings in people and their professions. If we were to live by the book, so much time and money would be wasted in correcting technically incorrect things, but things that don;t actually matter in the long run.

 

[happyfunball]

Yea, I would agree that LakeEarth's example of scientific mischief is EXTREMELY uncommon in science. Using Photoshop to completely remove data is rediculously bad BUT we do use photoshop to adjust contract and brightness and make our pictures look a little bit better.

 

[LakeEarth]

Having said that I have photoshopped gels, but only to remove dust marks, or to remove entire lanes (that were irrelevant to the discussion at the time)

But that's exactly what it was.

edit - It was a band that had nothing to do with what I was trying to show, so including it just would've made things more complicated and more work for both myself and the reader. So I got rid of it. I would've just re-ran the gel not including that part but I was running out of time.

- and don't worry, I already told myself to never do it again fearing going down a slippery slope, plus this article shows me that I have to strive to be better than others, plus other factors... next time I'll make the damn time!

 

[Mandark]

Of course there are areas of scientific research in which there are obvious financial conflicts of interest. I was going to post this link but forgot.

 

[mrkgoo]

But that's exactly what it was.

edit - It was a band that had nothing to do with what I was trying to show, so including it just would've made things more complicated and more work for both myself and the reader. So I got rid of it. I would've just re-ran the gel not including that part but I was running out of time.

- and don't worry, I already told myself to never do it again fearing going down a slippery slope, plus this article shows me that I have to strive to be better than others, plus other factors... next time I'll make the damn time!

Oh, maybe I misunderstood...I thought you had 3 bands in one lane, and photoshopped two out - that would be really bad - you're actually altering the experimental result. If you are removing entire lans from a gel of 3 lanes, then you're basically only showing certain results (ie. doing three experiments, but only showing the one that worked), which is probably not ideal, but acceptable (especially if the othe experiments are totally explainable, and completely irrelevant to the discussion at hand).

 

[LakeEarth]

What I did was wrong, but not that wrong. Don't worry, I would NOT purposefully change my results to make it look like my result was somethign that wasn't true.

Thanks for the hard time though, I kinda wanted to get yelled at for doing it but didn't want to tell anyone about it in my lab :lol

 

[Aruarian Reflection]

You know, I'm working at the NIH this summer, which is one of the largest medical research centers in the world. I was pretty surprised (and a bit disheartened) to see some MDs/PhDs screwing around during working hours. Of course, I don't know the full story, and they may have just been waiting for an experiment to finish running, but it was pretty weird to hear a couple of doctors browsing bittorrent sites and playing Family Guy clips :lol

 

[happyfunball]

SnowWolf that shit is all too common in science. We have lots of experiments with long incubation times that require you to find something to do in that time....most of the time we just bullshit or fuck around on the computer.

 

[LakeEarth]

That's my main problems. During the breaks where I should be reading journals and shit I'm ON HERE :lol

Damn you all!

 

[videogamer]

You know, I'm working at the NIH this summer, which is one of the largest medical research centers in the world. I was pretty surprised (and a bit disheartened) to see some MDs/PhDs screwing around during working hours. Of course, I don't know the full story, and they may have just been waiting for an experiment to finish running, but it was pretty weird to hear a couple of doctors browsing bittorrent sites and playing Family Guy clips :lol

Will you be at the main campus or the Bayview campus? I heard that (the main campus at least) it's a cramped paradise. Lots of incredible equipment and smart people with very little bench space.

So are you working for them or interning? I wanted to intern this summer, but the time isn't there. Another thing I heard is that a lot of people there are publication machines. *wide grin*

 

[Brannon]

For the 1/3 that fudge results to get the grant money, the other 2/3rds are balls to the wall trying to find the flaws that will discredit them so that they can get the fame and the glory and the prize and of course that sweet grant money.

I mean damn, they make fun of this almost all the time on those crappy Sci Fi channel movies; THIS AIN'T NEWS PEOPLE.

Well, except the fact that you can replace money and fame with dinosaurs and a megalomanical complex to rule the New World Order of whatever group of the day is trying to assume power but, well, yeah.

 

[mrkgoo]

That's my main problems. During the breaks where I should be reading journals and shit I'm ON HERE :lol

Damn you all!

Haha, sometimes, I should actually be doing work, but I'm here INSTEAD! Anyway, I was curious what it was like in other labs - is it basically come and go as you please, so long as you have good output, and the work that needs to be done is done? Or is it strict 9-5 stuff?

There's so much downtime in Science, but you're supposed to find constructive ways to fill in that time (paper-writing, literature research, or just coming up with new ideas and stuff - there's always stuff you could be doing that's not browsing GAF!).

 

[LakeEarth]

Yeah, for me it's come when you need, leave when you want kind of thing. My prof keeps teasing the other profs cause their students are usually gone by 5 while his stick around. Well, except for me, but I manage my time well and I'm usually completely done any experiments by then. Plus I'll stick around if I have to, I don't mind too much, but usually if I know I have a big day of work ahead of me I come in earlier rather than stay in later. So yeah, it differs from lab to lab.

My prof was telling me how once he went to some University in Tokyo or Osaka or something, and the 'big famous guy' came in at around 11am. My prof was shocked since this guy was really big in his field. Turns out the guy went home around 6am in the morning and was just coming back after a quick nap. And he did it EVERY DAY!

 

[BigGreenMat]

You know, I'm working at the NIH this summer, which is one of the largest medical research centers in the world. I was pretty surprised (and a bit disheartened) to see some MDs/PhDs screwing around during working hours. Of course, I don't know the full story, and they may have just been waiting for an experiment to finish running, but it was pretty weird to hear a couple of doctors browsing bittorrent sites and playing Family Guy clips :lol

Welcome to science buddy. I work in the field and there is more than a bit of screwing around, but for a lot of people they put in a LOT of work. I mean do you think scientists have to constantly stick their nose into every journal published and scour over their data every second?