Catch up on stories from the past week (and beyond) at the Slashdot story archive

 



Forgot your password?
typodupeerror
×
Science

Positive Bias Could Erode Public Trust In Science 408

ananyo writes "Evidence is mounting that research is riddled with positive bias. Left unchecked, the problem could erode public trust, argues Dan Sarewitz, a science policy expert, in a comment piece in Nature. The piece cites a number of findings, including a 2005 paper by John Ioannidis that was one of the first to bring the problem to light ('Why Most Published Research Findings Are False'). More recently, researchers at Amgen were able to confirm the results of only six of 53 'landmark studies' in preclinical cancer research (interesting comments on publishing methodology). While the problem has been most evident in biomedical research, Sarewitz argues that systematic error is now prevalent in 'any field that seeks to predict the behavior of complex systems — economics, ecology, environmental science, epidemiology and so on.' 'Nothing will corrode public trust more than a creeping awareness that scientists are unable to live up to the standards that they have set for themselves,' he adds. Do Slashdot readers perceive positive bias to be a problem? And if so, what practical steps can be taken to put things right?"
This discussion has been archived. No new comments can be posted.

Positive Bias Could Erode Public Trust In Science

Comments Filter:
  • by Gothmolly ( 148874 ) on Friday May 11, 2012 @07:51AM (#39965477)

    Right? isn't that what American schools and TV have been teaching for the last 30 years? Nerds aren't cool - facts are open to interpretation - everyone is special - you can eat more than you grow... When you have a society rewarding irrationality, what do you expect? Rigorous science?

    • by AwesomeMcgee ( 2437070 ) on Friday May 11, 2012 @08:19AM (#39965771)
      I wonder how many of these "positive bias" results come from the fact that if you publish results that disagree with the bias of those who are paying for the study, they'll probably ensure it's never published and you'll find yourself no longer running studies on their dollars.

      In the tech industry we all deal with non-technical managers who drive the technical direction and often times define the message to the clients. Does science suffer the same unskilled managerial types pushing scientists to interpret results in a particular way perhaps?

      I have a hard time believing a professional scientist doesn't know how to apply the scientific method, but then again incompetence is rampant in every other industry I guess, why not the scientific one..
      • by Geoffrey.landis ( 926948 ) on Friday May 11, 2012 @08:44AM (#39966075) Homepage

        Actually, science is stll working; the real trouble comes with the publicity of the science.

        You should never believe the results of any single study. Every scientist knows this; or should know this. Science comes when results are confirmed, not when somebody publishes the first paper. The real work of science just starts when somebody publishes a study saying "we show that x has the effect y." That initial paper really is no more than "here's a place to start looking." However, newspapers want to publish news, and they need to publish whatever's hot and interesting and being done today, not "well, scientist z had his team take a look at the xy phenomenon to see if there was anything interesting there, and they couldn't really find anything there, although maybe some other research lab might have different results."

        And, I suppose that somebody should post a link to the obligatory xkcd: http://xkcd.com/882/ [xkcd.com]

        • by silentcoder ( 1241496 ) on Friday May 11, 2012 @09:02AM (#39966277)

          There's a brilliant line on this in "The science of discworld" - I won't pretend I can quote it 100% accurately off the top of my head but it goes something like this:
          "In the media you will often read that a certain scientist is trying to prove a theory. Maybe it's because journalists are trained in journalism and don't know how science works or maybe it's because journalists are trained in journalism and don't care how science works - but a good scientist never tries to prove her theory, a good scientist tries her best to disprove her theory before somebody else does it for her, failing to disprove it is what makes a theory trustworthy."

          • by slew ( 2918 ) on Friday May 11, 2012 @10:07AM (#39967173)

            I think the real problem today, is that jounalists think they can be stars out of the gate. In the old days, you started out as a fact-checker, before you could get even get a word that you wrote published (usually anonymously at first). By the time you got a "by-line" you have come up through the trenches and seen all of the behind the scenes mistakes that the "star" writers made. Today you have a blog and no editor. Not only quality journalism goes out the window, but these new so-called journalists don't learn the consequences for some of their habits, because they haven't seen others make them (and haven't been motivated like a fact checker to find the problems).

            Of course this "star-at-birth" issue isn't just problem with journalists, but many professions (e.g., scientists, chefs, programmers, etc)...

          • And Feynman said as much in his famous "Cargo Cult Science" speech, which I encourage everyone to go read.

            The problem is a lot of scientists DON'T do science that way these days. Bad science, "positive bias" as this article calls it, was rampant in the behavioural sciences when I was studying them. I basically never read a paper where they falsified their theory, or where they said things were inconclusive. They always found a way to hammer the results in to support of their theory, and none of them were ev

        • Actually, you need to read "The Black Swan" by N. N. Taleb. Science that tries to confirm a theory is already infected with confirmation bias. There are a pile of examples that demonstrate the fallacy of confirmatory inference. Taleb uses a variant of Bertrand Russell's -- a turkey might reasonably infer, based on his daily experience, that humans exist for the sole purpose of feeding him, caring for him, providing for his every need. This might go on for day after day, increasing the turkey's degree of belief in his hypothesis of a good and beneficent humanity filled with love of turkeys, right up to the day that -- ulp -- something unexpected happens.

          I'm not a Popperite, rather a Jaynes-Cox-Bayesian, but nevertheless it is important to avoid confounding the relative strength of positive and negative evidence. Absence of evidence is not the same as evidence of absence, yet we almost invariably confound the two.

          Taleb damn skippy agrees with you about publicity, however, and the near-criminality of publicity and reporting of science. A newspaper necessarily takes a scientific result or observation and transforms it two ways: First of all, it creates a narrative. It isn't just "a tornado hit Houston", but "a tornado hit Houston, possibly caused by Anthropogenic Global Warming" with the subtext "this isn't an act of nature, random an unpredictable, but is instead our fault". Aztec priests couldn't have come up with a better excuse for ripping the still beating hearts out of a stream of slaves and war captives. Second, it necessarily reduces the complexity of the result to no more than three variables, ideally one. It "Platonifies" it (according to Taleb) -- wraps it up in a pretty, easy to understand package that makes it more predictable, less random than it really was. Global warming is a much simpler "cause" than "A cold front overrunning a warm wet surface layer of air near the ground, creating turbulent rolls that break off and terminate on the ground, sustained and driven by the thermal difference, and it is a better story too.

          Sadly, as you point out, real science is all too often (and should be) scientist z looked at something and didn't find much. But what they failed to find and how they looked is actually often as or more important than a study that claims to find something, especially when the latter uses questionable methodology to try to prove something, cherrypicks data (for the same purpose), ignores silent evidence (ditto) etc. Medical science is permeated with this. Nobody gets famous, or rich, or even a job, for looking for a cure for cancer and not finding one. This too is addressed by Taleb. Great book.

          rgb
      • by danbuter ( 2019760 ) on Friday May 11, 2012 @09:14AM (#39966445)
        Sadly, in many cases, it's other scientists that are causing the problems. Back when I was in college, my Geology professor was trying to publish a paper that would have invalidated the results of an older study. Unfortunately for him, the major Geology magazines all used a similar pool of professors who were "experts" on that particular topic. One of those reviewers was the geologist whose work was being overturned. Let's just say that my professor's work was shot down quite quickly. (He did get it published, but in a smaller magazine, that honestly has little impact in the field).
        • Let's just say that my professor's work was shot down quite quickly. (He did get it published, but in a smaller magazine, that honestly has little impact in the field).

          Juggs?

      • by next_ghost ( 1868792 ) on Friday May 11, 2012 @09:15AM (#39966463)

        Positive bias works a little differently than you think. First, a few definitions. Positive result looks like this: "We tried X on Y and it works." Negative result looks like this: "We tried X on Y and it doesn't work." Which one looks more interesting? That's right, the positive one. Now remember that outside mathematics, science is stuck with probabilities. There's always a very small chance of false positives (and also false negatives, but those are less of a problem).

        So let's suppose that we have an experiment which should come out negative with 99% certainty. There's 1% chance of false positive. Now let's have 100 scientists independently perform the experiment. Chances are that 1 scientist will get a false positive. This scientist will then publish a paper while the other 99 will give up and research something else without publishing because of the impression that negative results are not interesting.

        This is how positive bias works. Those 99 negative outcomes need to be reported to show that the one false positive is a false positive but they aren't reported. It's not some kind of intentional fraud but merely a consequence of overt obsession with original and sensational results dictated by those who pay for research. It doesn't matter what the research actually means, only that it can be phrased as "We tried X on Y and it works."

        • by radtea ( 464814 ) on Friday May 11, 2012 @09:50AM (#39966981)

          During most of my career in pure physics I got negative results, which were always hell to publish. One of the things you'd hear a lot was, "Don't worry, a negative result is just as good as a positive result!"

          At some point I started telling friends and colleagues who got positive results, "Don't worry, a positive result is just as good as a negative result!" Which is false, as proven by the fact that no one but me ever said it.

          Negative results are hard to get published, but far more common than positive results. Furthermore, on the road to any positive result there are going to be lots of negatives: even today, working in an area where true positives are much more common, I try to put a section in every paper entitled something like "Things That Did Not Work So Well", because any experiment or computation or theory is likely to involve some dead ends that seemed like a good idea at the time, and if scientists don't report on them they will continue to seem like good ideas to people who haven't tried them, who will then waste effort on trying them, and fail to publish them when they don't work...

        • by jpate ( 1356395 ) on Friday May 11, 2012 @09:56AM (#39967049) Homepage

          This is how positive bias works. Those 99 negative outcomes need to be reported to show that the one false positive is a false positive but they aren't reported.

          I just want to point out that it's not quite so straightforward. A null result for an effect is not, in and of itself, negative evidence for that effect, it's just a lack of evidence for that effect. It's always possible that a different set of materials, a larger sample size, an additional control, more sophisticated stats, or any number of methodological modifications would succeed in finding an effect. 99 null results with bad materials are not evidence against even a small number (not one!) of positive results with good materials. Null results are under-reported because they are much more ambiguous, not (only) because they are harder to sensationalize.

        • Actually, it is a bit worse than that. Let us assume that eating Twinkies has no affect whatsoever on toenail cancer (TNC) rates. Let us assume that 20 groups set out to measure the affect of Twinkies on TNC. Since "proof" levels are traditionally set at p=.0.05 levels of significance, it is quite possible that not one, but two groups will get "significant" results. One group proves that Twinkies cause TNC. The other that Twinkies prevent TNC. Both groups will try to publish. They will likely succeed

      • by matthewv789 ( 1803086 ) on Friday May 11, 2012 @10:35AM (#39967469)

        Yup, that's the crux of the problem. While it may be true, as others say below, that publication bias against negative results occurs in all fields (such as physics) regardless of study funding, what we are seeing now is the influence of pharmaceutical industry funding in the clinical trials used for FDA approval of drugs (that is, a company funding the trial of its own drug).

        Specifically, drug studies funded by pharmaceutical companies are four [newscientist.com] times [healio.com] more likely [latimes.com] to show a positive benefit than ones funded by neutral sources. This is a problem because nearly two-thirds of clinical trials used for FDA approval are now industry-funded.

    • by SirGarlon ( 845873 ) on Friday May 11, 2012 @08:29AM (#39965911)

      This is a dilemma that is really goes to the heart of the philosophy of government. If the majority is irrational, is it better to give them self-determination and accept they will make frequent bad decision, or have the enlightened few rule them and impose better-informed decisions upon them?

      Hint: there is no correct answer. I am not an historian but as far as I know this debate between a pure democracy and some form of republic goes back to Rome and Greece.

      What is kind of weird is that the two major parties in America have developed into philosophies that are kind of opposite their names: the Democrats favor the paternalistic nanny state governed by the enlightened few (what I would call a "republic"), and the Republicans favor the ignorant mob ("democracy").

      As an aside, when America was a young nation many of her leaders advocated public education as a way to narrow the gap between the elite and the general population. That does not seem to be working out real well, though.

      • by Rakishi ( 759894 ) on Friday May 11, 2012 @08:56AM (#39966221)

        Here's the thing, there are no enlightened few. There's just a few equally irrational people whose irrationality makes them think they are rational and all knowing.

      • by grep_rocks ( 1182831 ) on Friday May 11, 2012 @10:20AM (#39967327)
        Mod me down or whatever but I think you really got it wrong - one party is for an elite or aristocracy based on financial wealth, they spend a lot of money on ads TV networks etc. to get a majority to vote in favor of the aristocracy, lower taxes on the rich less restrictions on the use of capital, less labor laws, less environmental regulation - i.e. freeehdum! the other party is a bit more disorganized and is probably more easily defined as the not-moneyed elite, but they still have to produce leaders and they tend to be more technocratic - so for example one party puts an oil executive as head of the dept of energy, the other put in a nobel prize winning physicist
    • by ArcherB ( 796902 ) on Friday May 11, 2012 @08:31AM (#39965925) Journal

      Right? isn't that what American schools and TV have been teaching for the last 30 years? Nerds aren't cool - facts are open to interpretation - everyone is special - you can eat more than you grow... When you have a society rewarding irrationality, what do you expect? Rigorous science?

      Considering that I'm done growing, if I didn't eat more than I grow, I'd die of starvation.

      • Right? isn't that what American schools and TV have been teaching for the last 30 years? Nerds aren't cool - facts are open to interpretation - everyone is special - you can eat more than you grow... When you have a society rewarding irrationality, what do you expect? Rigorous science?

        Considering that I'm done growing, if I didn't eat more than I grow, I'd die of starvation.

        Interesting interpretation of what GP said. Not in context, but interesting. Obviously he meant eat more than you need to grow and sustain yourself, i.e., get fat -- and of course we can argue that getting fat is growing, just in the wrong direction; but again, that would miss his point.

    • by Belial6 ( 794905 ) on Friday May 11, 2012 @11:09AM (#39968007)
      The 'facts are open to interpretation seems to run like this: 1) Statements are either facts or opinions. 2) Facts are statements that are correct. 3) Thus all other statements are opinions. 4) Opinions can't be 'wrong' because they are not facts, and are inherently subjective. 5) Thus all Opinions are correct. 6) Statements that are correct are Facts. 7) Opinions are facts. This leads to: 1) Red is blue. 2) Red is blue is incorrect. 3) Since red is not blue, "Red is blue" is an opinion. 4) "Red is blue" can't be wrong because it is an opinion. 5) Thus "Red is blue" is correct. 6) Since "Red is blue" is correct, it is a fact. 7) Red is blue is a correct fact.
  • by Anonymous Coward on Friday May 11, 2012 @07:57AM (#39965525)

    Positive Bias is another word for Group Think. I guess it could also mean deception [bishop-hill.net]

    • Positive Bias is another word for Group Think. I guess it could also mean deception [bishop-hill.net]

      Replying so parent gets noticed despite a downmod 'Overrated' from score zero.
      Please read the link as the behavior it exposes is highly relevant to the article.

  • by atlasdropperofworlds ( 888683 ) on Friday May 11, 2012 @07:58AM (#39965539)

    There are "studies", and then there is observation, modelling, prediction, model testing which is this thing called science. "Studies" are bullshit. Scientific research functions as it should. I believe the OP's article is just a chunck of sensationalist BS, or utterly ignorant of what science is (and is not).

    • by jeffmeden ( 135043 ) on Friday May 11, 2012 @08:02AM (#39965569) Homepage Journal

      There are "studies", and then there is observation, modelling, prediction, model testing which is this thing called science. "Studies" are bullshit. Scientific research functions as it should. I believe the OP's article is just a chunck of sensationalist BS, or utterly ignorant of what science is (and is not).

      You forgot to mention that it is yet another piece of published work that suffers from positive bias...

    • by fearofcarpet ( 654438 ) on Friday May 11, 2012 @08:38AM (#39966007)

      There are "studies", and then there is observation, modelling, prediction, model testing which is this thing called science. "Studies" are bullshit. Scientific research functions as it should. I believe the OP's article is just a chunck of sensationalist BS, or utterly ignorant of what science is (and is not).

      That is not really what TFA is talking about. Daniel Sarewitz is re-phrasing a long-known problem with "studies," as you call them, which is that complex systems are--by definition--too complex to study as a whole. I am a physical scientist, which means that I typically make or measure something in a well-controlled experiment and then change variables in order to test a hypothesis. I can basically publish a paper that says "we tried really, really hard to find it, but it wasn't there." In the life sciences, they are trying to answer vague cause-effect questions like "does this drug affect a particular type of tumor more than a placebo." Thus researchers in those fields have to create models in which they can control variables. He gives the example of mouse models, which are obviously imperfect models for human physiology. How imperfect is the question. The creeping phenomenon that he is addressing is the tendency to relax the standards for what counts as positive evidence--and I'm grossly oversimplifying--by waving your hands around about how mouse models are imperfect, but that there is definitely "a statistically significant trend." The root cause is simply the ridiculous amount of pressure that life science researchers are under to publish, which requires results, because their methodology is standardized. Those poor bastards can spend eight years on a PhD project that goes nowhere or burn four years of their tenure clock figuring out that their experimental design was flawed. *Poof* no funding, no tenure, no degree, time to consider a new career. That sort of potential downside creates the sort of forced-optimism that TFA describes.

    • Both studies and science are affected by bias inducing forces, such as 'publish or perish' policies of institutions and grant availibilty from stakeholders with an economic interest in the results. Elsevier, et al, raise similar problems with their control of the distribution of knowledge pipelines.

      About a hundred years ago traffic problems became so bad that government had to step in and legislate which side of the road drivers had to use, and who had right of way at intersections. It might be that simila

  • Wait, what? (Score:2, Insightful)

    by bmo ( 77928 )

    'Nothing will corrode public trust more than a creeping awareness that scientists are unable to live up to the standards that they have set for themselves,' he adds.

    No, the corrosion of public trust is the incessant idiocy coming from Fox and other Murdoch properties exclaiming "oh those silly scientists got it wrong again!" when the story is about a refinement of a model or something.

    Scientists are losing the credibility war because scientists are not PR flacks and are unable to counteract the "we don't h

    • Re: (Score:3, Insightful)

      by Anonymous Coward

      Oh yes, blame it all on fox, the ultimate evil in the universe. It has nothing to do with studies being published making lavish claims which are then later proved false, or so wildly overblown that it's almost embarrassing. Of course, the conduct of the scientists themselves couldn't possibly be at fault and it must all be an even republican conspiracy.

      Grow up, pull your head out of your ass, and realize that fox and republicans aren't the only source of evil in the world.

    • by tp1024 ( 2409684 )

      Not being an American, I don't know what FOX is putting out. But if they are pointing out that any arbitrary number of "X causes cancer", "Y prevents cancer" and "Z causes heart disease" studies are bullshit, then it is not merely their perfect legal right to do so, but they are also delivering an accurate describtion of such "research".

      Just because somebody is a murderer and thus a bad man, doesn't mean he also picked your pocket, ran a red light and parked his car in front of your garage.

      • Re: (Score:3, Informative)

        by AlecC ( 512609 )

        The trouble is that most such "X causes cancer" statements come from the media themselves, recklessly shortening research results saying "consumption of X correlates with a positive increase in cancer", where the nature of the correlation is unknown and the increase is very small. So it is all to often not an accurate prediction of the research. Particularly, there can often be a chinese whispers effect, where the researcher publishes a paper, the University PR department publishes a precis edited for PR pu

    • Re:Wait, what? (Score:4, Insightful)

      by gandhi_2 ( 1108023 ) on Friday May 11, 2012 @08:16AM (#39965729) Homepage

      This is a story about actual bias in scientists which is affecting the quality of their research.

      And you completely gloss over that to take issue with Fox, which is no more ore less biased than msnbc, et al...
      Look, people seek an echo chamber. "News" companies of all types just supply the demand.

      I suppose YOU don't see a problem with some news organizations taking biased scientific output and unquestioningly running with it as though it were the concrete truth for ever more.

    • Re:Wait, what? (Score:5, Insightful)

      by thegreatemu ( 1457577 ) on Friday May 11, 2012 @08:31AM (#39965929)

      While I agree that models are frequently refined, leading to new results, there is a disturbing trend I see, not having to do with positive bias necessarily, but with uncertainty estimation.

      One thing that I've found incredibly hard to beat into undergrads taking my physics lab courses is that getting your uncertainties (or error bars) right is far more important than getting the right central value. This is because uncertainties are the only way that two experiments can be compared against each other, or the only way to compare experiment to theory. If I have two models of climate change, one of which predicts a temperature rise of 3 C ± 5% and another that predicts 4 C ± 7%, those results are in large disagreement, whereas two studies that predict 20 C ± 15% and 40 C plusmn 35% are in much closer agreement.

      But I see it seems much more frequently, especially in fields like astronomy, too little thought goes into the systematic uncertainties, and you'll get 4 experiments measuring the same thing with results that cannot be reconciled if you take their statistics at face value. This was a huge problem with many of the early global warming predictions as well; every year a new estimation would come out that was completely incompatible with the previous one. Yes, these models are insanely complicated, and it's damn hard to understand all the systematics. And of course you can't put in error bars for plain old mistakes. But do it too many times, and people begin to lose any faith that your estimates can be relied on for anything.

      This is the problem I see; not necessarily bias toward a positive result, but a bias toward underestimating the uncertainty of your measurement, which I suppose could be different sides of the same coin. (E.g., a result of 2 ± 0.1 is a positive result; a result of 2 ± 5 is not!).

    • by invid ( 163714 )
      In order to understand what this bias means, people have to understand the scientific process, and many people don't. Such people would be more convinced by scientists pontificating on high that they are always right, the way religious leaders do.
  • data point (Score:5, Insightful)

    by Anonymous Coward on Friday May 11, 2012 @08:01AM (#39965555)

    I received my PhD in physics, and the thesis was measuring a number, in which I measured zero within the error bar. Not particularly interesting, but valid science. My wife was in a PhD program in Biology, she also did valid science, novel measurement technique, came up with an uninteresting result, therefore was not able to publish, therefore was unable to graduate. It would have been extremely simple to fudge the result to a 2-3 sigma result 'hinting' at an interesting answer, which would have gotten published. I think certain sciences have gotten to a point where they have forgotten that if you do valid work in a novel way, then that is science and you should not be punished for the conclusion of the measurement. Most measurements you do of the natural world should probably end up being unsurprising, and thus uninteresting, but you don't graduate or get tenure with those kinds of results. I think this is the mechanism for the positive bias. That is why I do not take results from certain branches of science at face value.

    • Re:data point (Score:5, Insightful)

      by spiffmastercow ( 1001386 ) on Friday May 11, 2012 @08:13AM (#39965697)
      This. Until we start giving PhDs for finding expected results and/or verifying the results of others, we're going to have this problem of research 'exageration'.
      • Re:data point (Score:4, Interesting)

        by Hatta ( 162192 ) on Friday May 11, 2012 @08:44AM (#39966077) Journal

        Which is exactly what we need to do. We need an entire independent arm of the sciences dedicated to confirming established results. This is something that needs to occur separate from the pressure to come up with novel results to please grant reviewers.

        • by martas ( 1439879 )
          Actually there doesn't need to be an independent arm of any kind. All you need to do is get the government to allocate funding for grants dedicated to verification, of course with strict criteria regarding competence of the researcher in experimental protocol and statistics. And publish verification results in university or government run open-access journals.
    • It's valid science, but unless it conveys important information to other people, why would other people be interested in reading it in a journal?

      No one remembers the 100s of ways Edison found not to make a lightbulb - even though they were each as important as his one successful attempt.
      • Re: (Score:2, Insightful)

        by Anonymous Coward

        Original commenter here:

        It shouldn't be in the highest tier journals, but should be able to be published somewhere so people can find the technique and that someone had done the measurement, with result x. And if the idea and methodology is sound, it shouldn't prevent someone from getting a PhD. But it does. And it filters out a lot of people who would choose to publish results that do not further their own career, as well as filtering in people willing to do things like playing with how they cut their d

      • Re: (Score:3, Insightful)

        If it's valid good science it speaks to the competency of the individual who practiced it, if we aren't graduating these folks, then we're encouraging graduation of incompetent or sensationalist scientists.
        • Agreed.

          The science should be of a publishable quality, but that doesn't always mean the result is publishable. AC's wife got screwed.
    • Re:data point (Score:4, Interesting)

      by fearofcarpet ( 654438 ) on Friday May 11, 2012 @08:51AM (#39966153)

      I spent some time in the Chemistry department at a fancy ivy leave school that seemed designed to crush the spirits of biologists. While we were off drinking beer and playing volleyball against the Physics department, they were toiling night and day on multi-year projects that may or may not generate publishable results. The sick thing is that the outcome was totally decoupled from the abilities of the researchers--it was more like a test of stamina and luck. I saw talented, brilliant people turn into bitter husks. Some gave up on research and wound up teaching at private colleges, others left science completely--and not into fields that used any of their expertise. It seems like a lottery where those that get lucky (or stab the most backs) go on to academic positions at top-tens and the rest end up in rocking back and forth in the corner mumbling to themselves all day.

  • by tp1024 ( 2409684 ) on Friday May 11, 2012 @08:02AM (#39965561)

    The solution lies in a reformation of research finance that is not focussed on how many papers X published compared to Y, but also takes into account whether they are consequential or not and if they actually comply with at least basic scientific attributes such as repeatibility, verifiablity, falsifiability, accessibility of all data and all conducted research, as well as actually conducted verification of research by independent third parties.

    There should also be an outright condemnation of data mining, where data bases are checked only for the existence of attributes and correlations that happen to affirm the researchers opinion and leave all others untouched.

    Fields like economics, medicine and climate have long since deteriorated to mere cargo cults due to those failings.

    • by rwv ( 1636355 )

      So would you have scientists publish fewer original research papers and more papers that attempt to reaffirm or disaffirm research that's been published by their peers? I'd be surprised if there isn't *some* resource that links research publications with a list of secondary papers that "Support the Same Conclusion" and "Refute the Conclusion". Given that scientists are looking to publish (churn out) a lot of papers... it seems low-hanging fruit would be studying and trying to repeat the Research/Conclusio

      • by tp1024 ( 2409684 )

        Even if there were such lists, they are not what research funding is based on.

        What funding is based on is which journal published your article and how many citations it received in other papers - whether the author citing your paper has actually so much as read (much less made use of) your paper or not. Just being popular with peers helps, because you can always find some excuse or other to cite a paper even if it is semi-relevant at best.

    • by slimme ( 84675 )

      Data mining is indeed a very mediocre scientific activity. Correlation on itself means nothing at all. If you want to proof something the correlation should be 100% and you should be able to explain why the correlation exists and replicate it in controlled experiments. The problem is that those slam dunk scientific discoveries are all or mostly allready found. And nowadays the poor scientists need to find something to bolster their path to glory.

      Good science could be: find a correlation an proof the causali

    • by ceoyoyo ( 59147 ) on Friday May 11, 2012 @08:44AM (#39966069)

      "also takes into account whether they are consequential or not"

      That makes the problem worse. We need to evaluate papers based on whether they are good science or not, and not publish the ones that are bad science. Currently the "negative results" which are actually inconclusive results, are not published because they are, well, inconclusive. Unfortunately a lot of the "positive" results are also inconclusive, but they ARE published. The solution is not to publish more "negative" results, it's to stop publishing the flawed "positive" ones.

  • by Life2Short ( 593815 ) on Friday May 11, 2012 @08:02AM (#39965563)
    Some sort of redirection towards findings that can be verified by independent labs would seem to be an improvement on the current system. But that would require a focus non science as a system and less of the "great researcher" emphasis we see today.
  • It also erodes science.

  • by Karmashock ( 2415832 ) on Friday May 11, 2012 @08:12AM (#39965681)

    They're overstating precision. Which rather then a forgivable error is an elementary mistake no trained scientist should ever make.

    A VERY basic concept they teach at the lowest level of science education is the distinction between accuracy and precision. This is science 101.

    Accuracy is whether or not a given conclusion is correct.
    Precision is to the degree of specificity.

    Typically you run into problems on complex subjects because they overstate the precision of their data or their ability analyze the data.

    This can boil down to simple thinks like significant digits.

    For example, I'm measuring volume to two significant digits in a giant data set with thousands of measurements. When and if I average those numbers the final average can't have more then two significant digits. That sounds elementary but you see this error made on some big studies. You'll have a situation where something is being measured in a crude sense by many sources and then in the analysis a much higher degree of specificity is implied.

    Often that degree of specificity is required to make certain conclusions which is why they break the rule. This is lazy and a breach of scientific ethics. What they need to do is collect the data all over again this time to the level of specificity they need.

    Simply saying its too hard to collect the data properly so they're going to make assumptions is not reasonable or ethical. I suppose you could do it so long as you kept an asterisk next to the data and the findings to make it very clear throughout that the conclusion is a guess and not in any way empirical science since at some point people were guesstimating results.

    • by medcalf ( 68293 )
      That's one of the problems I have with global warming. The precision of the best made, best sites thermometers is about a degree. One common class is +/- 5 degrees. Yet these are being used to derive results in tenths of a degree. Satellite proxy measures help, but that's a very short record in climate-significant time frames. Is the Earth warming? Almost certainly. Is that outside the norm? Hell, we don't even know if it's outside the margin of error.
      • Well, it's funny when they do that with proxy data. They use samples of biomass in sediment from 10,000 years ago and pretend to derive a .01 temperature difference. You can't get that from pine cones, ice cores, and fossilized vegetation.

        The real problem with all this stuff is that the data was never collected for this purpose. The standards and precision were for crop forecasts and whether or not people should go to the beach. It was never collected to be compared to a tenth of a degree across continents

    • by ceoyoyo ( 59147 ) on Friday May 11, 2012 @08:50AM (#39966139)

      "When and if I average those numbers the final average can't have more then two significant digits."

      Yes, the average can have more precision than the individual measurements. That's actually kind of the point of an average. It can't improve accuracy though, for the most common definitions of accuracy.

      Your definitions of accuracy and precision are sort of right, but also sort of misleading. And the way you use specificity is incorrect. But as you correctly point out, a lot of working scientists are a bit fuzzy on all these concepts too.

  • I used to be in academia, and it was no secret that researchers almost always found the results that they had planned to find from the beginning in their studies. I don't think I ever once saw a case where a researcher started out with a hypothesis, found it was completely wrong through the research, and then let it go. There are a million ways to cook the numbers to reach the conclusion that you want to (and the one that gets you the grant money and publication). If a certain position is popular (and well

  • Apophenia (Score:4, Insightful)

    by concealment ( 2447304 ) on Friday May 11, 2012 @08:16AM (#39965725) Homepage Journal

    When you have a hammer, everything looks like a nail.

    Anything can become a religion, as a result. We're less critical of our data when that happens, and we "nudge" it into place.

    The problem is not "science" per se but our social approach to it.

  • I would say that the bias is not so much positive as it is "get noticed and get more grant money" biased.
    Unfortunately, you get noticed more and get more grant money if you find what you were looking for instead of disproving your initial hypothesis.

    Also this article seems to imply that the problem that needs to be fixed is that of public trust, while I would argue that the public should distrust a community that gets it wrong so often (it is the skeptical, scientific thing to do). Unfortunately, far too ma

  • Like just about everything else in this world, science is about money. And how do you get money in science? By finding and/or hyping the next leap forward. Being successful in science is all about getting grants. You don't get tenure without bringing in grant money, you don't get grant money without publishing in the best journals, you don't publish in the best journals without finding the next leap. Your typical PhD finishes school in their late 20's, probably with significant school loan debt. He o
  • by l00sr ( 266426 ) on Friday May 11, 2012 @08:24AM (#39965831)

    In engineering research, there is definitely a positive bias; in fact, negative results are rarely published at all. This is both because negative results have less sex appeal than positive results and because peer reviewers are trained to outright reject publications without positive results. Although there is huge pressure to publish positive results, I'm not aware of systemic fraud in the literature. What does happen, however, is roughly this: 1) researcher gets great idea. 2) researcher tries idea. 3) idea fails to produce state-of-the-art results. 4) researcher adds hacks and kludges to marginally improve performance. 5) repeat steps 2-5. So, what you get in the end are journals filled with "positive results" that mean nothing and a bunch of "scientists" who make a living doing things that do not really resemble science at all.

    • by ceoyoyo ( 59147 )

      I bet what you think are negative results are actually inconclusive results. LOTS of people, including many of the ones writing papers about positive publication bias, make that mistake. An insignificant p-value is NOT a negative result. It's an inconclusive one. In order to actually show a negative result you have to do more work, and delve into the (usually very simple) stats that few people know. Most studies don't have the power to show actual negative results, but in my experience if you actually

  • I'm not sure if it's just me, but I've got the feeling that these days, most useful and ground breaking new research and technology comes from companies, not universities.

    What do you think?

    • by ceoyoyo ( 59147 )

      It's just you. Corporate R&D produces products. Sometimes incremental engineering improvements. Not science.

      Corporations USED to do some decent science, even more or less basic science. But not anymore.

    • by geekoid ( 135745 )

      I think that's a classic keyhole problem.

      Private companies have a shit ton of marketing and PR.
      90% of all projects in large companies fail. 10% succeed. The marketing never mentions the failure.

      In the government, 90% of all project are successfully, 10% fail. Because the PR(news media) only focuses on the negative, the success don't get talked about.

      This is the same thing.

      .

  • Publish or perish. And journals prioritize first results on a research question, just like newspapers want to scoop the competition. On top of that, even first news on a hypothesis is less likely to be published if it's negative.

    So what do you expect academics to do? Reinvestigating a previously reported research result is unlikely to get funded, it's unlikely to be published even if it is researched. They live and die by publications.

    Unless and until the publication system makes it possible for aca

  • by codegen ( 103601 ) on Friday May 11, 2012 @08:30AM (#39965921) Journal
    The case of the green jelly beans.. http://xkcd.com/882/ [xkcd.com]
  • by elsurexiste ( 1758620 ) on Friday May 11, 2012 @08:44AM (#39966067) Journal

    I remember a few days ago someone submitted a story about piracy for "The Avengers" being low compared to potential profits from them. A few high-ranked comments were like "This is yet another proof that [insert common /. parlance here]". I saw very few comments that stated the most plausible reason: a camcorded action film, with crappy audio and a shaking image, can't compete against the real thing. I thought the same thing: confirmation bias.

    People do it all the time. If something can somehow support their views (specially if they don't RTFA) they'll use it as yet more confirmation. "I still don't get why this piece of evidence is discarded by everyone else! They must be delusional or have bad intentions". For example, I imagine this article will be used as evidence for: lack of funding, falling standards in the US, the demise of education, lack of scientific reasoning (maybe they'll even extend it to scientists themselves), and other common /. utterances. I wonder how many of them will actually say what I found out after RTFA...

    So, everyone is playing the same game, and scientists are no exception. But hey, that study has numbers on it. At least you can try to replicate the findings, if only the entry barrier wasn't so high: these tests are *hugely* expensive. More collaboration may be a good idea. Shared laurels are better than none, right?

    P.S., a nice article on confirmation bias (and other goodies) here [youarenotsosmart.com].

  • by ceoyoyo ( 59147 ) on Friday May 11, 2012 @09:13AM (#39966423)

    Okay, it may not be exactly a myth. We can't tell. I strongly suspect there is actually a negative publication bias.

    What most people think are "negative" results are actually inconclusive. A non-significant p-value is NOT a negative result. That misunderstanding is very widespread, and leads to lots of high level mistakes. Half of the neuroscience papers published in top journals including Nature the last two years that could make a mistake based on that fallacy, did. And neuroscience didn't seem to be particularly worse than most other fields.

    A non-significant p-value is just that - not significant. Inconclusive. Getting an actual negative result is considerably more work than getting a positive one. You need to figure out what the minimum effect size you're interested in is (you should do that for positive results too, but almost everyone just uses zero for that) and show that your confidence intervals do not include it. As Ionnadis points out, you really should consider the power of your study as well (also for positive results), and take a stab at estimating the priors too.

    If you go and do all that, and also do a quality experiment, in my experience you actually have a pretty good chance of getting published, because a) it's clear to the reviewers you've done a really thorough job (any idiot can run some data through a t-test and get a p-value, negative results are harder) and to show a negative result your study is probably much higher powered than a positive result one, meaning an impressively big p-value.

    The problem is not that there's a positive publication bias, it's that most scientists don't know how to show negative results so there are very few negative papers around.

  • by Goldsmith ( 561202 ) on Friday May 11, 2012 @09:45AM (#39966905)

    I'm a scientist. It's a big problem.

    Here's how you fix it:
    The metric for success for both researchers and their government funding sources is published papers. It's hard to change the cultural view of the scientific community, but it's easy to change government metrics. Paper publishing as a metric is easy to track between programs, but has had a terrible impact on scientific culture. It's also led to a large bias in how the government decides what areas to fund. If your metric is paper publishing and you're looking at energy issues, do you fund a sub-field with a historic high paper publication rate, a moderate paper publication rate or a low publication rate? It's fine for us here to say we'd fund the best research, but a government program manager may lose his job for picking a field with the lower publishing rate.

    Other metrics such as how many other researchers use some results or whether a practical implementation of some new technique is developed will be harder to judge and take a longer time to evaluate, but would at least give us an honest assessment of the quality of government funded research. Tie future funding to what our broader society is looking for out of science, and eventually the scientific culture will follow.

  • Let's all try to be careful, here.

    Medical research was a little late the science party, and they still have serious issues to work out. Some of these problems can't be helped (low sample sizes lead to higher p values, but higher sample sizes place more human beings at risk), others can (not being skeptical enough of research conducted by organizations with a financial stake in the outcomes).

    Most of these problems are peculiar to medical research and should not be conflated with all of science. While we should hold the medical research community's feet to the fire over this, the reason we are hearing so much about this is that they themselves are talking about it and they themselves are conducting the meta-research that is producing all of these articles. Other branches of science have looked down on medical research for a long time, but at least they are now getting serious about addressing the problems.

  • by tgibbs ( 83782 ) on Friday May 11, 2012 @11:39AM (#39968407)

    Something that most scientists know, but that is not widely appreciated by the public, is that "landmark" studies are particularly subject to positive bias. For a study to be acclaimed as a "landmark," it is the first report of a novel phenomenon, which by definition means that it has not yet been confirmed by other investigators, and moreover that it is in some sense unexpected--which means that there is not even much "indirect" evidence from other sources that it is correct. It is also likely to be submitted to one of the high-profile "newsy" journals, like Science or Nature, where submissions are not evaluated solely on the basis of whether the science is high quality, but are first screened (and usually by an editor, before it even gets specialist peer reviewers) on the basis of whether it is of "broad interest." Because these journals are widely read and cited , they have high impact factors [wikipedia.org], so even getting published in one of these journals is a feather in a researcher's cap. In contrast, a publication in a middle-rank journal does not give a major immediate boost to a scientist's career, but may have a long-term benefit--but only if it turns out to be correct. Being "scooped" (by having somebody else publish research leading to the same conclusion) will not prevent you from being published in a middle-rank journal, but it will knock you out of the newsy journals. So the alluring prospect of publication in a newsy journal may lead a scientist to be a bit less cautious than normal, and the risk of being "scooped" means that he/she may not take usual precautions (like seeing if somebody else in the lab can reproduce the result independently). High profile journals are stringently reviewed, but peer reviewers even in these highly-regarded journals necessarily take the authors at their word, unless there are blatant errors--nobody visits the authors' labs to look over the shoulders of the researchers or to verify that negative results have not been discarded. On the other hand, if you are planning to publish a result in a middle-rank journal, you are more likely to take the time to make sure that your conclusions will hold up over the long haul.

    Working scientists know this, so they take these "landmark" studies with a grain of salt until they are independently confirmed. But the public can easily confuse "high profile" with "most reliable."

To stay youthful, stay useful.

Working...