Science Has a Reproducibility Crisis

Edward Ferrara writes:

If your Facebook feed is anything like mine, you may have recently heard about how Bill Nye–the Science Guy himself–“slammed” Tucker Carlson on the latter’s evening show on Fox. THIS. (If you live somewhere else you may have been treated to an equally smug reaction from people claiming that Carlson “won.”)

However you feel about it, the timing, coupled with Nye’s reliance on scientific consensus as a proxy for objective correctness, is somewhat serendipitous. Mounting evidence that the results of scientific studies are often not replicable has caused Nature, one of the most prolific scientific journals, to very publicly tighten its standards for submissions as of its latest issue.

In May of 2016, a survey by Nature revealed that over two thirds of researchers surveyed had tried and failed to reproduce the results of another scientist’s study. Over half of them had been unable to reproduce their own results. Fifty two percent of researchers polled said there was a “significant crisis” of reproducibility.

This is a big deal. The ability to replicate the results of studies is crucial to both scientific integrity and progress. Clinical researchers, for example, depend on reliable results from prior trials to form the building blocks of new drug advancements. In the field of cancer biology, merely 10% of results from published literature were found to be reproducible. Meanwhile, the credibility of scientific literature is understandably compromised by dubious, often sensational findings.

The root of the problem, according to Dame Ottoline Leyser, director of the Sainsbury Laboratory at the University of Cambridge, stems from today’s scientific culture. As quoted in BBC, she cites “a culture that promotes impact over substance, flashy findings over the dull, confirmatory work that most of science is about.”

Others blame a pressure to publish. There has also been, in recent years, doubt cast on the integrity of the peer review process, especially with regard to climate science. Whatever the culprit, plans to combat issues of reproducibility are emerging. Nature has developed a checklist to serve as guidelines for authors submitting writing to the publication. Efforts shouldn’t end there, the journal argues. Reform at all levels of the scientific process could go a long way:

“Renewed attention to reporting and transparency is a small step. Much bigger underlying issues contribute to the problem, and are beyond the reach of journals alone. Too few biologists receive adequate training in statistics and other quantitative aspects of their subject. Mentoring of young scientists on matters of rigour and transparency is inconsistent at best. In academia, the ever increasing pressures to publish and chase funds provide little incentive to pursue studies and publish results that contradict or confirm previous papers. Those who document the validity or irreproducibility of a published piece of work seldom get a welcome from journals and funders, even as money and effort are wasted on false assumptions.

Tackling these issues is a long-term endeavour that will require the commitment of funders, institutions, researchers and publishers. It is encouraging that NIH institutes have led community discussions on this topic and are considering their own recommendations. We urge others to take note of these and of our initiatives, and do whatever they can to improve research reproducibility.”



182 thoughts on “Science Has a Reproducibility Crisis

  1. If you can’t reproduce the result , it isn’t science.
    That should be tattooed on every scientists wrist , where it is impossible to hide.

      • Either you can reproduce the results of whatever work you have undertaken to show there was a Big Bang, or there wasn’t one (as far as science is concerned).

        An event in and of itself isn’t the science, (I walked to the shop today isn’t science…), it’s the way we show that it happened.

      • Actually, if you can’t reproduce observing the CMB then the Big Bang theory would be unscientific.
        But you can.

      • Greg F. Quite right. By it’s nature an hypotheses cannot be reproduced, it can only be tested. In most cases, with yet more hypotheses. This is, for me, the problem with AGW: it’s still only an hypothesis, yet the supporters of it would like the public to think that it’s a proven fact. Unfortunately, they don’t like the public to see what they consider proven facts so they hide their ‘workings’.

        The public need to be reminded, climate change is a proven fact; AGW is not.

      • I was surprised but glad to see the mention of BB and CMB. The BB theory and how it is talked about almost everywhere it is talked about is as if it is fact, like CAGW. It takes some time but it is possible to put ‘inflation’ and ‘inflatons’ firmly in to the science fantasy catagory and not a very robust one at that – tooooo ‘invented/contrived’. Just to mention that ‘inflation’ requires physics that goes against everything that forms the bedrock physics of today. (not that that is such a strong argument admittedly) CMB is rather questionable too.
        I just wanted to join the voice that is seldom heard about the ridiculous idea that is the BB – I would question whether it is a theory even. ‘Inflation’ is a definitive leap of the creative imagination to well beyond the plausable just to satisfy a different, rather shaky conclusion about CMB.
        (not to mention singularity)
        I’m surprised BB doesn’t get a laugh/snigger every time it is mentioned.

      • Actually, sorry, no Big Bang, not that it is irreproducible but because it is based on a baseless assumption that all red-shifted light is due to receding stars and completely ignores, except when talking about black holes, that gravity can red-shirt light as well. Everything is red-shifted because everything has gravity.

        They basically took the math of a theoretical Big Bang and kludged their astronomical findings to match their models. Their prediction of light elemental abundances and the background temperature of the universe were both very wrong until they magically created a fix in their math model.

        The Big Bang model is so bandaged that it is a joke. Along the way, to explain the fact that the model called for more mass than was detectable, they created Dark Matter, the boogey man in the closet, and then created A Dark Force and Dark Energy to support this fallacy. Having created a Dark Physics which supposedly covers about 95% of the universe, 95% we cannot detect, they want more money to study this magical realm of sheer fantasy.

        If you go back to the Steady State Universe that was dominant until a bunch of physicists jumped, for not good reason, on the Big Bang wagon, virtually everything we can see or detect is easily explained using normal physics, including plasma physics. No need to dream up an entirely fake world of Dark Stuff that no one can detect, “But, we need more money because detecting the non-existent takes a long time and a lot of money.” That’s job security only equalled by the climate “scientists” studying warming that is no happening.

      • Scientific investigation begins with theory. The “Big Bang” is theoretical, so is anthropogenic global warming.

      • The “Big Bang” is more of an explanation. It’s the explanation that best fits the currently known facts.
        New facts or a new explanation could over turn it at any time.

      • Charles Higley, Why would you expect the universe to be full of visible matter. It is big and so dark. Light spreads out dimly.

        The Cosmic Microwave Background radiation is evidence for the Big Bang. It is the echo, murmuring in the twilight.

        The fact is that, wherever we look in the universe, the laws of Gravity seem to be the same. They may not be explicable but they are uniform. Why should that be if the universe wasn’t all in one place when the laws came into being?

        The Big Bang theory has lot of supporting evidence for it. It has passed from hypothesis to reliable theory.
        Although, I admit that Dark Energy is a bit embarrassing.

      • Big Bang, was result of Bean eating feast at Hogwarts, and arrant spark from Harry Potter’s wand. Obviously…

      • Charles Higley:

        There also materials that affect the propagation of light. The “Big Bang” theory is restricted by assumptions about the character and behavior of intervening “space”. It’s the same limitation which precludes prediction of climate change on Earth. It’s part of the chaos (i.e. incompletely and, in fact, insufficiently characterized and unwieldy) that created a need to define and constrain the scientific domain. So-called post-normal science and scientists have adopted a “secular” philosophy, faith, or fantasy that indulges in inference (i.e. created knowledge) and estimates (e.g. models) for political progress (e.g. to marginalize or suppress competing interests. for secular profit).

      • “Big Bang theory has lot of supporting evidence for it. It has passed from hypothesis to reliable theory”

        Ehhh, No.
        No explanation as to where that single point came from.
        No explanation as to why that single point suddenly went Big Bang. According to Big Bang WAG, time didn’t exist prior too, thus nothing advances.
        No explanation as to where nothingness around single point came from.
        No explanation as to where required anti-matter went.

      • Leo Smith

        Oh yea? Did you see it? Lots of stuff happened so that when you extend the picture backwards, you appear to get a big bang…but you also get a lot of infinities. Kinda like the center of a black hole. Which mans we (and, certainly you) do not fully understand it.

        Instead of using you fairytale to discredit “reproducibility”, you better first understand your data (which is what this is all about).

      • I hold firmly to the “I have no ******* idea what happened” theory of the origins of the physical universe.

      • Jbird on March 6, 2017 at 7:29 am
        “Scientific investigation begins with theory. The “Big Bang” is theoretical, so is anthropogenic global warming.”

        Not true. Almost all scientific investigation starts with an observation. Then a hypothesis (an educated guess) is formed about why or how the observed thing occurred. And then tests, experiments and protocols are established that are supposed to be designed to FALSIFY that very hypothesis, not to prove it. The goal is to attempt to disprove your hypothesis every single way you possibly can. If you can’t falsify it, you publish it, and let others replicate your experiments, and try any additional ones they can think of. If no one can falsify your hypothesis, it becomes a valid theory.

        That doesn’t necessarily make it a scientific law, or even true in every situation.

        The problem with AGW is that someone took a known reaction (that bottles filled with high concentrations of CO2 respond differently to radiation than those same bottles filled with regular air) and extrapolated that reaction across the whole world’s atmosphere as if they had already run all of the falsifications required to justify calling it a valid theory.

      • Leo Smith, you say (cynically, if I interpret your comment correctly): “Once (and only once) upon a time there was a Big Bang”. Einstein conjectured what it would be like to ride on a beam of light. So, let’s look inside the sun at a single nuclear explosion, and conjecture what it would be like to be inside it, a moment later. I think it would look just like our universe looks now, and that you could analyse what was going on around you to work out when your own BB occurred, just as scientists are doing now in our universe. So maybe our universe is just one explosion in a universe of explosions, ie, there are as many universes as there are nuclear explosions in stars in this one universe that we can see. And there is no need to stop the logic at a singlle iteration, upwards or downwards. Einstein changed how we see space-time. Is there another change coming?

      • >>
        Charles Higley
        March 6, 2017 at 7:14 am

        Everything is red-shifted because everything has gravity.

        Not everything is red-shifted. M-31 (the 31st object in Messier’s catalog; also known as the Andromeda Galaxy; the Great Andromeda Nebula, or the Great Spiral Galaxy in the constellation of Andromeda) has a negative red-shift or is blue-shifted. It would make sense as she’s the bigger sister to our Milky Way Galaxy in the Local Group or Cluster, and we’re coupled by gravity. Her distance is only about 780 kiloparsecs (give-or-take 20 kiloparsecs), and our combined closure rate is about 300 km/s. They expect a collision between our galaxies in about 4.5 billion years or so.

        Along the way, to explain the fact that the model called for more mass than was detectable, they created Dark Matter . . . .

        Well, not exactly. It’s whether or not you think Kepler’s third law is valid. Objects orbiting closer to the center of mass have higher orbital velocities than objects further away. Since stars orbit the center of gravity in a spiral galaxy, their orbital velocities should decrease as you get further away from the center of mass. Observations show that those stars have constant orbital velocities from close in out to nearly the edge of those galaxies. That either means that our physics is wrong (a point that may have some validity), or there is a lot more mass than can be seen.

        If you go back to the Steady State Universe that was dominant until a bunch of physicists jumped, for not good reason, on the Big Bang wagon . . . .

        Which cannot explain the CMB. The so-called Big Bang can. When plotted, the points from this radiation form a nearly perfect black-body curve of about 2.7 Kelvin. The “bunch of physicists” jumped because there was no good reason to stay with the Steady State Universe. The CMB is actually not the Big Bang itself, but a period about 380,000 years after when hydrogen was cool enough to become transparent. Above a certain temperature, hydrogen becomes opaque–it’s one of the reasons why stars, like the Sun, are not transparent.

        No need to dream up an entirely fake world of Dark Stuff that no one can detect,

        That is a problem, of course. But if you believe in standard candles, and if you believe that type IA supernovae fit that bill, then you need to explain why the overall expansion of the Universe is speeding up.


      • I’m with Dems B. Dcvrs March 6, 2017 at 8:23 am.

        The ‘big bang’ is absolutely no explanation of anything.

        What there before that?
        What was this ‘point’?
        How was it created?
        Why did it ‘bang’?

        It is the atheists’ version of the ‘god’ explanation.
        (And we thought our minds had escaped that particular ‘explanation’.

      • FG – I think the point was the opposite – if you can’t reproduce it then it’s not science. I can reproduce simple tunes on the piano but nobody ever said that means my piano playing can be described as science.

      • Then Forest, by all means, tell us what “science” is to you? (You’re not a post-normal scientist, are you?)

      • Science is the use of the scientific method and the body of knowledge produced by that method.

        There are numerous explanations. Some of the most accessible are by Richard Feynman “It doesn’t matter how beautiful your theory is, it doesn’t matter how smart you are. If it doesn’t agree with experiment, it’s wrong”.

      • Trying to falsify is doing science. Reproducibility is simply trying the same falsification testing as the original. This is good practice, because the original experimenters may have selected a non-representative sample, they may have tainted their own tests accidentally (or intentionally, it’s been known to happen). The repetition may expose a design weakness in the experiment or mal-intent by the original experimenters.

        What is NOT science is trying to PROVE something in the prosecutorial sense, picking a desired conclusion (humans burning fossil fuels will catastrophically heat the Earth’s atmosphere) and then making a case for it, looking for proof that it is true. This is where CAGW post-normal science has gone totally off the rails. Now they’ve expanded their sophistry to say, if we can make a computer model that approximates the appearance of our atmosphere, that therefore the premise must be true. To me, this is not only NOT persuasive, it’s getting close to mental illness, intentionally creating a mass delusion.

      • All science should work the same. I should be able to take you data, your methods and your experimental setup if applicable and I should be able to reproduce your results. That’s the first step. Another step would be to confirm your results with either slightly different methods or experimental setup. Once you’ve got to that stage. The you’re probably onto something. But the first stage of confirmation is independent self consistency. Should that fail then you need to reexamine both your work and your conclusions. Remember it’s not sin to be wrong or to have reached a dubious conclusion. It is a sin to stick your fingers in your ears and shout lalalala…. better still start calling your opponents deniers and you’re good to go….. please don’t keep calling it science though as it does tend to annoy the rest of us :-)

      • But observational sciences like geology, astronomy, etc. do not allow experimentation, so the reproducibility criterion doesn’t apply to them.

      • In those cases I have a set of observations. I interpret those and for instance make some predictions. If those predictions hold true. Then maybe I’m in to something. Astronomy is like that in general. I develop a theory about condensed matter. I also further predict the existence of magnetars and I either sit back and wait for someone to actually find one or I go looking for it myself. There is a school of thought that merely making and collecting observation is not science. It’s data collection. Many of our climate science brethren often do a lot of this. They then start applying statistical analysis ocaisionally without fiddling with the source data and it’s at this point that the science theoretically starts. In other words the science normally begins when you try to interpret your observations. Hope that helps

      • Critically, I should show my workings so that others could reproduce my work. Consistently coming to the same or at least similar conclusions. You could look to say evolution or relativity Based on observational interpretations these were just theories until science caught up. Similarly quantum mechanics. If you want to have more than a theory or an idea. You have to be able to test it or hope that someone else will be able to test it in the future. Only then does it truly become science rather than speculation. Even then its a continous process. A single experiment could for instance prove you wrong. It could alter a consensus and lead to a paradigm shift. At least that’s the way it should work.

      • Observational sciences like astronomy, geology, and climatology have very little room for experimentation, so they are scientific even though findings in those fields can’t be reproduced.

      • Sorry Roger but that’s not true at all. Geology: I predict there is a cap rock here and I predict say the presence of oil or gas. You drill. You find oil or gas. Theory partially confirmed. I take another location make a similar prediction you drill.. . Nothing…. I go away I refine my theories. I make new prediction you drill etc. Etc. Theories get refined by being tested they are reproducible or they aren’t and they are discarded. Without being able to tell me more than you could by sheer luck it’s not really science it’s data collection, curation or sample collection.

      • Roger

        So reproducibility does not apply to observational sciences (i.e. astronomy)…

        Then why do astronomers reproduce the gravitational bending of light at every solar eclipse (or the transit of Mercury)?

        Agreed there appears to be a continuum or reproducability. This fact is the back door thru which psychology and economics try to justify their activities as “science”.

        I guess this discussion demonstrates the need for many of us to go back and read (or re-read) Kuun. The lack of so-called “scientific discipline” being tolerated/encouraged by academia today is a path back to the dark ages.

      • Okay. Say we just talk about climate science and we disregard any data fiddling. I have a collection of data. Yay well done what can you tell me about it? I can tell you it was 12 degrees C at midday on the 21st of May 1998 at location Y. Useful historically but not much else. I take lots of this historical data. I take what I understand about the climate, circulation, radiative physics, thermodynamics, particulates, boundary conditions, state changes, albedo etc etc and I build a model… all good so far…. I try a bit of hindcasting… I tweak my model…we go through a few iterations… I get to a point where that seems to work alright. In other words I develop my theories and interpretations. Having done all of this i switch into predictive mode. I use the models to try and tell me something useful about the future. With me so far?

      • things that were once called natural history ..

        At this rate, one day economics will be deemed science, after all the science part is already tacked onto the back of the course name :/

        Botany, Zoology and other natural history type courses are often taught in the science faculty as well they should be, they use many of the tools of science (things once known as ‘philosophical instruments’) which brings me back to the term ‘scientist’ – a term coined by a group of philosophers who sought to separate themselves from other philosophies by using a different method of logic.. rather than using substantive inductive logic – find supporting evidence of things, they used falsification – trying to disprove things. To this end science is a very limited tool with very limited application.

        Too many popular science personalities have tried to suggest science can be used more broadly to solve problems without comprehending that science isn’t the tool… Those other methods of logic actually are legitimate and do have applications and it’s almost a shame science became the grand logical method – it’s caused others who use elements of science or tools of science to think they’re using science.

        It brings to mind the story of Einstein addressing a medical conference explaining ‘you dress like scientists, you talk like scientists, but you are not scientists’. Science is a great tool, but too many think anyone with a degree in science is a scientist. Too many cannot tell the difference between real science and something that looks and sounds like science. If being a scientist only meant having a degree in sciences, then the consensus on eugenics in the 20’s must have been right.. after all, all the scientists agreed. If science is a philosophy then those supporters of that foul theory were wrong because they failed to falsify it and chose to gather supporting evidence instead.

      • Roger Knights-
        “But observational sciences like geology, astronomy, etc. do not allow experimentation, so the reproducibility criterion doesn’t apply to them.”

        Excuse me, but geology and astronomy are both PHYSICAL sciences. In both, there are multiple kinds of research, including both observational studies and experimental studies. The only reason we successfully put men on the moon was because we conducted decades of physical and chemical experiments here on earth first. We didn’t just “observe” birds flying and think….”we can get there with really big wings”. NASA’s scientists had to reproduce weightlessness, study it’s effects and come up with ways to counteract it. They had to reproduce centrifugal forces and study their effects and come up with ways to compensate for them. Same thing with gravitational forces, and the effect of tides, and how the minerals and materials in different soils react to physical forces etc. And the lives of every astronaut we’ve ever sent into space has depended upon the absolute ability of scientists to get the same result every single time.

        Observational studies are just those done by watching something and collecting data without human interference in the study. Experimental studies are those done in which human control the protocols to see how doing so affects the results. And most sciences utilize BOTH in order to learn about our physical world.

      • Aphan, putting men into space and onto the moon was more an engineering task than a scientific feat. The engineers used science to accomplish these things, and the engineers deserve more credit than the scientists.

      • “Blunderbunny March 6, 2017 at 9:40 am
        Sorry Roger but that’s not true at all. Geology: I predict there is a cap rock here and I predict say the presence of oil or gas. You drill. You find oil or gas. Theory partially confirmed. I take another location make a similar prediction you drill.. . Nothing…. I go away I refine my theories. I make new prediction you drill etc. Etc. Theories get refined by being tested they are reproducible or they aren’t and they are discarded. Without being able to tell me more than you could by sheer luck it’s not really science it’s data collection, curation or sample collection.”

        You are conflating observation and prediction processes.

        You predict a cap rock: Based on what?
        Perhaps you’ve conducted a number of tests and you identify echoes that you believe are from a cap rock.
        That is your analysis and prediction.
        It is not a science theory.
        Nor are you reproducing an experiment. You could be simulating a prior experiment.
        You are following existing theory and practice.
        That does not make your analysis or prediction a new science theory.

        The same goes for believing the cap rock holds a reservoir of carbon compounds or water.

        You are correct that the physical sciences are subject to the same strict standards regarding reproducible results.
        e.g. Paragenesis regarding the formations of almost any body of mineral or ore emplacement undergoes frequent re-analysis and reframing the processes. All are theories, some are testable. All of the testable theories must be reproducible to confirm the theory.

      • ATheoK. I was sloppily taking the observations themselves as read because he was claiming that observational science was inherently different to other branches of science and I was apparently ineloquently trying to set him straight. Also. Didn’t want to go down the ology type argument as that often offends… plus as I hopefully demonstrated its an argument that i dont subscribe to

      • From Wikipedia on the scientific method:

        Main article: Scientific method
        The scientific method seeks to explain the events of nature in a reproducible way.[nb 11] An explanatory thought experiment or hypothesis is put forward as explanation using principles such as parsimony (also known as “Occam’s Razor”) and are generally expected to seek consilience—fitting well with other accepted facts related to the phenomena.[1] This new explanation is used to make falsifiable predictions that are testable by experiment or observation. The predictions are to be posted before a confirming experiment or observation is sought, as proof that no tampering has occurred. Disproof of a prediction is evidence of progress.[nb 12][nb 13] This is done partly through observation of natural phenomena, but also through experimentation that tries to simulate natural events under controlled conditions as appropriate to the discipline (in the observational sciences, such as astronomy or geology, a predicted observation might take the place of a controlled experiment).

      • Hi Roger.

        Yes that’s what I was trying to explain to explain. The tests maybe different, but the process is the same. Those tests shpuld be reproducible. If they are not then your hypothesis is incorrect. And should be looked at again. There got there in the end :-)

    • Not necessarily. Science cannot progress without tests that may fail. Failure to reproduce a result means nothing more than something was wrong. It might be theory, the model, the design or conduct of either experiment or oodles of other possibilities. It certainly is science, the very essence of science.

      What isn’t science is ignoring experimental results that aren’t consistent with previous experiments..

      • If Roger Knight thinks that astronomers and geologists don’t do controlled experiments, he’s misinformed. And that particular quote from wikipedia under Scientific Method regarding “observational sciences” is unattributed, so it could have been written by anyone. That someone said it, and RK agrees with it, does not make it fact, just like David Dirkse’s comment on who deserves more credit for putting men on the moon is an opinion, rather than a fact.

    • It is easier than that (tattooed scientists); disbelieve anyone claiming to be ‘scientist’. I know of meteorologists and geologists and physicists and physicians and psychologists, but of no one that is a scientist.

      Believe nothing read or heard without verifying it oneself unless it congruent to ones preexisting Weltanschauung.

    • The skill of the experimenter plays a part in the outcome though. Just because an inept experimenter cannot reproduce the result, does not mean that the result is invalid.

      There is also the issue that some experiments by their nature cannot be repeated multiple times under the same conditions. This is a particular problem with any experiments where randomness plays a part. In fact this is one of the greatest issues with establishing anything definite about climate.

    • There is laboratory science and there is forensic science. Forensic science, by its very nature, tries to use scientific instruments and known physical/chemical/biological behavior to reconstruct something that went on in the past. Forensic science cannot exist without laboratory science, because the instrumentation and physical models on which the scientist depends must be tested and verified in a laboratory environment. So, yes, a one time event can’t possibly be repeated. However, the consequences of that single event can be measured in time, space, perspective, etc. as many times as we like and the past reconstructed provided that there is a unique solution.

  2. ‘a culture that promotes impact over substance,’
    This was the main takeaway from Climategate for me. From it stemmed the correspondents’ obsession with the ‘message’.

  3. It’s time to settled the Scientific Method to a Reproducible facts and have a new words for the unreproducible one. My English is not good enough to clear it but we are moving from Aristotelian Binary Logic to Zadeh Fuzzy Logic (
    This means that we will have no more True and False as “Truth degree” but a continuous gap from 0.0 to 1.0. So we can have an half truth (0.5). We can’t say this is a “scientific” but only “logical truth”!

      • great link, thanks for that. I’ve read a few of Poppers books (and Feynman’s too) but that’s a nice page to send to people who need an introduction to Popper without frightening them too much

    • The CAGW community is already living in that “reality”. Not only does each GCM give a different result (irreproducible), but also each run of the same GCM gives a different result. They have solved this dilemma by creating an “average of GCM results” and assigning some “truth value” to it. Balderdash!

      • The problem is that Popper defined pseudoscience as metaphysic. Fuzzy Logic isn’t a metaphysic. It’s a way to achieve a right conclusion even with uncertain data.

      • fuzzy logic is fuzzy, not logic.
        boolean values are not numbers. there is no spectrum. true/false is a binary alternative.

  4. Quote: Tackling these issues is a long-term endeavour that will require the commitment of funders, institutions, researchers and publishers.

    Yes it will.

    The next issue is the scientific method. Reports of its demise are not necessarily exaggerated.

    • I’ve realised that the so called “scientific method” refers to something that is common place in engineering – that is not assuming theory is right but making sure something works in practice. And as such, whilst all engineers just adopt this way of working … checking theory actually works in practice is something that academics have to be strongly even forcefully encouraged to do.

      Or to put it another way … the scientific method to academics is a bit like doing a tax return … it seems totally unnecessary to any honest person as you know you’re not going to diddle the tax … so why should you have all the work of filling out the form?

      • I can’t say I agree SS, unless you are characterising what scientific malpractice has become.

        The scientific method isn’t something which can be retrofitted. It is the method of enquiry used by people who can rightfully claim to be scientists.

      • Engineering calculations are often wonderfully complex and hard to understand but they are based on objective observance of physical events. It seems like much of climate “science” is doing things backwards.

      • @SS:

        Very well put. And, BTW, why I rank Engineers ahead of mere “scientists” in competency rank…

        Engineered goods must actually work…

  5. I think a large cause of this problem was a movement around the time of Thatcher to introduce a “payment by results” culture to academia. As a result, the focus moved to the quantity of papers published rather than the quality.

    And as a result, groups of academics started colluding with each other to help each other boost their “published paper” status so that they could get on. We see this in the number of people who are often cited as “authors” of a paper – they are clearly not authors, instead they are just people like department heads who can demand their name gets on – or academic scratching each other’s back so they both get ahead in the “performance” targets of numbers of papers published.

    From there is was merely a slight change, from publishing totally bogus papers, or as occurred … getting papers published purely to overturn contrary decisions made on Wikipedia pages – or ” pause buster” papers, or “moonlanding” papers.

    • Couldn’t agree more SS. The ‘publish or perish’ and ‘quantity over quality’ ethos pervading academia these days is absolutely disastrous and for very obvious reasons which do not require a PhD in order to recognise.

  6. Now we are in a post-scientific Era. The problem is that Politicians take place before Logicians. This is the real modern problem. The fact is that Politic try to take place of the past Catholic Church prior the scientific Era.

  7. I just want to know if this oncoming revitalization of the scientific method will be calculated into the social cost of CO2.

  8. Perhaps each undergraduate student should be compelled to undertake a replication study, and write up the results? Without being able to reference the replication (whether it succeeded or not) prevent them from public funding in the future.

    People would be queuing up to replicate…

    Researchers who have demonstrated they produce work that fails replication multiple times by multiple researchers cease to be eligible to future funding.

    You can see the future, papers written with exquisite care including archived data to enable replication, students learn the benefits of open, reproducible science.

    A win win.

    • “each undergraduate student should be compelled to undertake a replication study,”
      “People would be queuing up to replicate…”

      I remember going to those replication party’s when I was a student, Happy days !!
      Sadly, People are not queuing up to replicate with me anymore (:-((

      “Researchers who have demonstrated they produce work that fails replication multiple times by multiple researchers cease to be eligible to future funding.”
      – That would cut funding costs by 50% or in ‘climate science’ 97%.

    • I remember from my days that even in highschool it was the way we were taught, mind you that was in the 60’s and 70’s and a non public school. That may have something to do with it. Also those were the days that through examinations and assessments we were already being pointed into directions that we were “good” in from vocational to university and everything in between., very good system.

    • Then by definition it is not science, but is nonetheless true. The logical domains: science, philosophy, fantasy, and faith, intersect, where real events may originate in any one of them, and crossover into the scientific domain with improved perception (e.g. proximity, knowledge).

    • Thank you for that link. As someone who falls into several of the subject groups I found it most enlightening.

    • @roving, A HUGE thanks for that article, I fall into some of those groups as well and this really helps my search for answers. GREAT link !

  9. When a study can’t be reproduced one must ask a few more questions:
    1. What was the point of it, is it not wasted research? Money down the drain.
    2. Money that should’ve been spent on good studies by meticulous researchers. The same meticulous researchers who are out of a job because the cowboys published more studies.
    3. What about the time wasted. The time of everyone involved in publishing, peer reviewing, reading, and trying to reproduce, or apply the results?

    Irreproducible studies mark scientific regression, not progress.

    • It’s not wasted research. It shows the original study flawed and leads to finding why the theory or hypothesis is flawed. There’s no learning in science without trying hypotheses and seeing if they work. Maybe scientists need to be a bit more sure before running an expensive study, but there is much to be learned when a study fails.

    • As I see the problem, when a researcher gets a surprising result his study gets published. But given the nature of sampling, at a 95% confidence interval five percent of samples will produce an a “surprising” result, i.e., an “incorrect” one per the confidence sought. Hence, the studies that tends to get published (because of their surprising results) tend to be the incorrect ones, i.e., the irreproducible ones. Ironically, since 2.5 percent of the sampled results lie in each tail of the actual distribution two similarly designed studies could conceivably come up with “surprising” results in opposite directions from the norm (i.e., the true result.)

      So, I see surprising results as being: a) most likely to be published, and b) most likely to be found later to be irreproducible. My conclusion: Whenever a surprising result is found, the researcher himself should logically attempt to reproduce it. And, upon publication, the publisher should make every effort to encourage others to attempt to reproduce it. That seems to be what “Nature” is advocating. And, obviously, for that to be possible researchers publishing surprising results should make all data and techniques available to others. Failure to do so should result in denial of publishing altogether.

      And, yes, that would likely doom the global warming campaign.

  10. What we are seeing as science now is simply the triumph of credentialism over competence.

  11. Publishing is one problem, but press releases and poor science journalism in the stuff most people read (how many non-scientists read Nature?) is a far bigger problem because then the public and policy makers act on what is increasingly “false science”.

    Far too much science cannot be reproduced because it has been produced to fit a political need or view or agenda.

    • I think this is the critical point. You have to publish your findings precisely so that others can replicate or refute them, but when the original publication gets advertised and repeated across the media then any attempt to replicate is stamped on as being ridiculous.

      It is the misunderstanding of what a peer-reviewed published paper actually represents that is the basis of most of the problems. Peer review is simply three scientists in the field checking for glaring errors by reading the paper and (hopefully) the critical references looking for inconsistency. It is not a reproduction of the work nor is it an endorsement of the results – at best it is a break from publishing overly broad conclusions derived from those results

      The establishment pushing the one-sided view of the catastrophic nature of anthropogenic climate change might not be the only cause, but by giving “peer review” a mythical standing (while at the same time undermining it by replacing it with pal review and pressuring editors to toe the line) the popular impression is that a published paper cannot be questioned.

      This is the long-term damage done to the practice of science by scientivists.

  12. What we’ve seen in the climate science arena over the past fifteen years has
    been an echo chamber with one “conclusion” from one study bouncing around
    and landing in five or six derivative studies. If there’s faulty data or poor
    statistical handling in the primary study, it gets harder to spot or eliminate by
    the time the fifth or sixth “daughter” study is published.

    Reproducibility has been replaced by cutting and pasting with friendly
    peer-reviewed oversight and lots of jawboning.

    “Why should I give you the original data when you’re going to use it to
    find something wrong ?” pretty much ties a bow around the problem.

    • Indeed. I’ve lost count of the number of climate related papers I’ve read which begin the introduction with a sentence along the lines of “It is a well known/understood/characterised fact that carbon dioxide causes …” and from the outset the reader is left wondering exactly where the supposed fact has ever been demonstrated. For me the whole subject has become identical in its evolution with theology.

  13. “each undergraduate student should be compelled to undertake a replication study,”
    “People would be queuing up to replicate…”

    I remember going to those replication party’s when I was a student, Happy days !!
    Sadly, People are not queuing up to replicate with me anymore (:-((

    “Researchers who have demonstrated they produce work that fails replication multiple times by multiple researchers cease to be eligible to future funding.”
    – That would cut funding costs by 50% or in ‘climate science’ 97%.

    • There has been a tradition, since 1921, in organic synthesis to publish in the journal (peer reviewed) Organic Syntheses
      known to all purveyors of organic chemistry as Org Syn. As an organic chemist I am proud of this tradition reproducibility among my fellow organikers.


  14. A lovely idea. How many journals will enforce this? Same way as we’ve seen them fail to enforce their rules on archiving of data used in published works.
    Negative studies will still rarely see the light of day.

  15. A open honest scientific attitude?

    We have 25 or so years invested in the work. Why should I make the data available to you, when your aim is to try to find something wrong with it Dr. Phil Jones, Director of the Climate Research Unit at East Anglia University, email to Warwick Hughes, 2004

    “I’m getting hassled by a couple of people to release the CRU station temperature data. Don’t any of you three tell anybody that the UK has a Freedom of Information Act.” —Dr. Phil Jones, Director of the Climatic Research Unit, disclosed Climategate e-mail, Feb. 21, 2005

    “I’ve just completed Mike’s [Mann] Nature trick of adding in the real temps to each series for the last 20 years (i.e. from 1981 onwards) and from 1961 for Keith’s [Briffa] to hide the decline.” —Dr. Phil Jones, Director of the Climatic Research Unit, disclosed Climategate e-mail, Nov. 16, 1999

    “You might want to check with the IPCC Bureau. I’ve been told that IPCC is above national FOI Acts. One way to cover yourself and all those working in AR5 [the upcoming IPCC Fifth Assessment Report] would be to delete all e-mails at the end of the process. Hard to do, as not everybody will remember it.” —Dr. Phil Jones, Director of the Climatic Research Unit, on avoiding Freedom of Information requirements, disclosed Climategate e-mail, May 12, 2009

    It isn’t just the papers that are a problem.

  16. This is a well known phenomenon in the soft sciences and drug research. The expense of some of the modern science makes reproduction difficult as well. In the final analysis it is the honesty of the scientists which counts, as well as the realisation that the unexpected non-result can give rise to a seismic change on occasions. Michelson Morley comes to mind.

    • No no no NO! That is entirely the wrong approach! The more expensive the experiment the more beholden the designers are to make absolutely sure it is repeatable or otherwise all of that colossal expense is for nothing. Why do you think they have two hugely expensive independent detectors, ATLAS and CMS, at CERN or the twin LIGO detectors at different US sites or multiple massive water Cherenkov neutrino detectors on opposite sides of the globe?

  17. Well, a lot of the work presented here has a reproducibility problem.

    [most of your comments suffer from that same problem too -mod]

    • So you put WUWT on the same level as Nature?
      I guess you think journals have already gone the way of other legacy media.

    • At least I make criticisms under my real name. Exactly how were you moderating the discussion there? I wasn’t breaking any rules, unless you’d like to point me and the transgression in my previous post that requires moderator intervention?

    • Nye’s ludicrous logical inconsistencies were amusing tho. Don’t know why Carlson didn’t pick up on them.

      • Tucker’s field of study is history. Since much of history is a confusing mish-mash of “he-said-she-said”, logical inconsistencies wouldn’t necessarily ring alarm bells for him. Had Nye been facing another engineer, he might have fared even worse than he did.

  18. You may be right. My daughter is an Infectious Disease Doctor doing AIDS research, specifically trying to develop a treatment for infants born w AIDS. She has gotten some work published, but some of her work has been rejected by ‘major’ medical journals because (at least according to some of the reviewers) – it does not break enough ground to merit publication – it’s too ‘incremental’

    • That sounds believable. Surely the answer is to seek publication in less prestigious journals?

      The best journals want the most important advances (also, conicidentally, the advances most likely to be wrong). The lesser journals just want advances. Incremental and accurate is just what they are looking for.

      Are there “less prestigious” infectious disease journals?

    • That’s too bad, sometimes we only see the effectiveness of an approach by incrementally exploring the boundaries. Sometimes and incremental report provides the information that sparks the inspiration for the next breakthrough, but only if it is seen by the right (meaning a large number) of people in a field. Sometimes those incremental reports provide the definition of the hard boundary edge of an approach and prevents others from pursuing in that direction.

  19. … the ever increasing pressures to publish and chase funds …

    Because of the amount of money involved it is unsurprising that drug research is corrupt and unreliable. link

    There is also vile and nasty activist and social justice warrior pressure to not publish results that contradict their narrative.

    Galileo’s Middle Finger … is (Alice) Dreger’s account of the battle between activists and scientists that resulted in not only the chilling of academic freedom of speech, but also the hounding of prominent sexology researchers, including Michael Bailey and Ray Blanchard. Galileo is required reading for anyone who wants to understand a crucial piece of the history behind transgender activism (seen through the prism of Dreger’s experiences as an advocate for the intersex community). Dreger, who tends to speak (and write) her mind, recently resigned from Northwestern, having had quite enough of censorship from her medical school’s dean.

    Roger Pielke Jr. no longer researches anything related to the climate because of relentless pressure. link

  20. The pendulum swings back. If we have done our jobs properly the rising generation of scientists will see the hubris of thier elders. That will keep the flame of honest inquiry fed. It is always difficult to fight the zietgiest but we have managed to pull it off.

    Wonder why some can so readily condem the influence of prestige and money on medical science and dismiss its effect on climate science. Climate science is the poster child for agenda driven abuse of science.

  21. Science is not even in the same ball park as engineering when it comes to traceability and reproducibility, heck the spell-checker must be a scientist, it thinks these words don’t exist. Since the customer for science is usually the govt, which is the Mother of all useless customers, this fact alone probably explains why science gets away with it. A major culture shift is required before science can get its act together.

    • Extra-universal, perhaps universal? An entity (i.e. coherent) that is not constrained to human perceptive and causal limits? Maybe.

  22. Anyone claiming Bill Nye “won” is relying on his readership not having seen the exchange with Tucker Carlson. Similarly, I have encountered global warming advocates who give links to articles that in no way actually reinforce their point, again relying on someone not bothering to check their assertion.

  23. A lot has to do with the ways Universities attract funding, which often leads scientists into directions they don’t want to follow for the sake of their careers. The nature of Universities has changed a lot in the last 50 or so years as they have become more exposed to market and political pressures in order to obtain funding.

    • Do you think Universities have just become smarter at obtaining funding?

      I recently walked the grounds of the U of MN and was in awe of all the new buildings that have cropped up in 20 years. It was crazy. There are buildings where I didn’t think it was possible to put a building.

  24. “The root of the problem, according to Dame Ottoline Leyser, director of the Sainsbury Laboratory at the University of Cambridge, stems from today’s scientific culture.”

    I think the plant anatomy is a bit mixed up there.

  25. The fractious mainstream global warming debate is primarily focused on the magnitude of climate sensitivity to increasing atmospheric CO2 (“ECS” or similar).

    Since “CO2 lags temperature at all measured time scales*” (References 1, 2 and 3 below), this mainstream debate can be stated as: “BY HOW MUCH CAN THE FUTURE CAUSE THE PAST?”.

    My proven observation* is still at the periphery of climate science, yet it is the fatal flaw in BOTH sides of the global warming debate.

    On Reproducibility:

    There is NO evidence that increasing atmospheric CO2 has any significant impact on global temperature.

    The global cooling that occurred from ~1940 to ~1975, during the time that fossil fuel combustion (and allegedly atmospheric CO2 concentration) strongly accelerated, essentially DISPROVES the catastrophic humanmade global warming (“CAGW”) hypothesis.

    Consider IF we had a similar situation starting about now:
    Hypothetically, let’s say from 2020 to 2055 there was continued fossil fuel combustion and a significant increase in atmospheric CO2, and yet average global temperature cooled by ~0.5C.
    What would this say about ECS and the CAGW hypo? It would say that ECS is near ZERO and that the CAGW hypo is false.

    But this has already happened, which is why the warmists have “adjusted” the temperature record in order to minimize this ~35-year past cooling.

    In fact, we already have a good bound on the magnitude of ECS, which is NEAR ZERO, and strong evidence that the CAGW hypothesis is false. Do we need another 35 years of data to reproduce this same result? We don’t.

    For clarity, I reiterate my central hypothesis, for which there is strong evidence:

    My hypo does not preclude other major possible major drivers of CO2, such as deforestation, fossil fuel combustion, etc. They just do not significantly impact global warming.

    Furthermore, there is strong evidence that increasing atmospheric CO2 is beneficial to humanity and the environment. Atmospheric CO2 is not dangerously high, it is in fact dangerously low for the continued survival of carbon-based life on Earth.

    There is strong evidence that “global warming” is the most expensive false scientific crisis in history, and strong evidence of academic and criminal misconduct by its leading warmist proponents.

    Here is my challenge: Prove my central hypothesis wrong, with real evidence.
    (Practitioners of arm-waving, strawman arguments (e.g. the “Mass Balance Argument”) , appeals to authority and other such nonsense need not apply.)

    Regards to all, Allan :-)

    * References:
    1. MacRae, 2008
    3. Humlum et al, January 2013


  26. Bill Nye knows nothing of Climate Change Science. A Limerick.

    The Roman Northamptonshire wine
    was good, not excessively fine.
    So it just goes to show
    that Bill Nye does not know
    of Climate Change past, that’s my line.

    During the Roman warm period wine grapes were grown almost up to the Hadrian Wall, The the dark ages came and grapes no longer ripened in England. During the Medieval Warm Period there was at least one cheese farm on Greenland “Gården under sanden”, abandoned as the glaciers regrew, starting the “Little Ice Age”. We are still recovering from the little ice age. 2016 may have been a warm year, but most years since the ice age were warmer. See Chart.Greenlandgisp-last-10000-newWe are still in the sweet spot of a remarkable stable Climate, only more CO2 will save us from a new Ice Age.

  27. I ain’t no scientist, but from where I sit, as a science consumer, it DOES seem logical that, while experiments that ultimately fail are a necessary part of the science building process (isn’t that the definition of ‘experiment’?) any ultimate conclusions drawn must be testable, provable and reliable. For Nature magazine, what is the point in publishing a ‘conclusive’ study that cannot be verified by others? Thanks, but no thanks.

  28. “If you live somewhere else you may have been treated to an equally smug reaction from people claiming that Carlson won.”

    I’ve seen the interview on FOX as no American (German). And I found that the claim that Tucker Carlson had “cornered” Bill NYE was really true. In addition to stereotyped assertions that science is solidified and what agony we should endure in the future because of our ignorance (above all in terms of sea level rise), nothing came of Bill Nye. On the contrary, for me, Tucker Carlson scored with the conclusion that the climate is warming up, but it is not clear what proportion humanity has in it. This is a very reasonable point of view from Tucker Carlson, who also appeared much more naturally than Bill Nye, who wanted to be loose, but has mainly a pinched facial expression. Bill Nye refused to provide any information about the exact proportion of man in the warming of the climate, with the mention of the consensus that still exists. I would say: point victory for Tucker Carlson. Asked if Bill Nye had said that people who would not agree with his view, were thrown into jail, he departed with usual nebulous tactics. He seems to be afraid of Pres. Trump and its reaction. Under Clinton he would probably have repeated this dirt. His main point that climate warming would be catastrophic in the future was the rise in the sea level, in the course of which countries such as the North of France and the UK would partly sink into the sea. A terrible interview with an unthinkable Nye. If all the supporters of the AGW are so, then good night. The statement “Bill Nye the Science Guy believes that climate change skeptics suffer from” cognitive dissonance “, which prevents them from accepting how global warming affects their lives” speaks for itself. If Bill Nye is inside as much as he looks outside, he does not change anymore. In this context, it is becoming increasingly clear what blessing the victory trumps was for the skeptical community. He imagined in the best case, how such a dried-up cardinal of the AGW would appear under Clinton.

  29. I never saw any posts or memes about nye “winning” that segment. And I watched the segment when it happened. Nobody “won” that, because nye said nothing of any importance at all. He just kept hammering away at his talking points, completely ignoring what Tucker was asking him.
    The best part of Tucker Carlson’s show is that these morons keep coming on that are apparently completely unaware of what they’re stepping into. Tucker openly mocks them and laughs at them, and they demonstrate their own unfamiliarity with things they claim to know about.

    There’s never going to be a single event that convinces people that the “climate change” meme is irreparably damaged, but a lot of little events can make people realize what’s been going on. This was one of those, and was one of the few times lately that I openly laughed while watching TV.

  30. Reproducibility, done well, can still lead to simply replicating a type 1 or 2 error.

    Robustness is a better measure. Through this method confounding factors are more likely to be discovered, thus calling into question previous results. And it is the only way forward to establishing new paradigms of scientific theories. Robustness is the process of, and when done with due diligence, rejecting or not the proposed hypothesis. Climate Science has underfunded this important work and leaves open the real risk of the populous being taken down the primrose path.

    • Even so, if one doesn’t provide the data, observations, methodology, etc., reproducibility can not be attempted.

      Presumably, if enough other scientists attempt the reproduction, it will take only one to show the error.

      It appears that CAGW folks put forth as “climate science” fails on all counts.

    • A particular problem in climate science is the practice of calling model runs “experiments.” The results of these “experiments” are completely reproducible. This is a link to multiple studies agreeing that the contribution of natural variation to recent temperature increases is about 0%. They are all based on GCM’s or post hoc statistical models.

  31. The reason for NON REPRODUCIBILITY is simply in the numbers. When you have tens of thousands of Phds it is natural that the quality of the Phd is debased. Like in any other commodity of which there are too many.

    Tightening up this or that will make no difference. Once a currency is debased only time and the reduction of the numbers of Phds will help.

  32. No shit, Sherlock, as we say in the real world of stuff that has to work to sell, outside of the temples of academe.

    I know I said I’d go away but this load of unresolvable hot air is helping create a letter to Nature. The
    conversation has now reached the level of the communications in the Guard post of the Death Star Detention block, and the whole thing deserves a similar fate for the time it wastes.. IMO.

    The problem here is simple. Climate science just isn’t. End of.

    A lot of what Nature publishes is not science – as a physicist or engineer understands it, that produces laws that can be independently validated in repeatable experiments and used to make or predict things that work every time, as predicted. Shame NASA didn’t stick to real science, a dark day for its credibility. Trump may be right about that.

    Examples of pseudo scieces are, ipso facto: Climate science, biology, plate techtonics, sociology, psychoanalysis and economics, and cosmology. All worthy of study, to gather data and knowledge, perhaps help us understand them, BUT so complex, unrepeatable, multi variate and non linear in nature that any predictions based on emperical dats and history are simply probablistic guesses, no better than intuitive speculation, however well informed, not a basis for any response. In fact you might as well predict earthquakes as climate change. Some will come, soon, or not. Feynman also pointed this failing out, in terms of how a loosely defined hypothesis is inevitably unverifiable hence infinitely debateable, because its hypothesis cannot be tested, and the subject system is unable to meet the fundamentals of proof – but I can’t find the interview its in…anyone?

    e.g.: Hypothesis: More CO2 will create a biological response that reduces warming, by a negative feedback effect that increases energy radiated into space, caused by biological feedback not included in the undersubscribed guesses of climate models, or through some other plausible Gaia mechanism I can hypothesise and make the data fit a model of with my modelers’ “adjustments”. I’ll never have to prove it, but can’t get such an assertion peer reviewed or published by the climate inquisition, of course. Some one should do that though :-).

    Can’t be hard given how easy the climationist fraud has been to impose over reason and available reality by the simplistic assertions of its leading priests in journals like Nature. Glaciers building in some places, stable in others, retreating in others, as they always have, polar bears still there and becoming a pest,. Oceans barely moving in height, temperatures not subject to any dramatic positive feedback hockey Schtick that tracks CO2 levels, that we are all still waiting for. The opposite happened of course, in relative terms. The multiplier was turned down to adjust for that, etc. You can’t make it up. That’s climate modellers’ jobs.

    All these disciplines have merit, but are not “science”, they prove no laws under the strict regimes of the hard science of things we can measure and repeat. Climate change models , and all the other things, fit Feynman’s pseudo science definition. Just a self evident truth. The so-called climate science may be useful in responding to crises when there is no proven science, but proves no laws and cannot be repeated, because the “nature – al” experiments are unrepeatable, with variables not under investigation held constant.

    In particular these sciences offer policy makers no basis for massive expense to try to change an effect no one understands in detail, not happening as advertised, and in fact a minor problem compared to natural changes in history involving forces far greater than we can conceive of, but that we should prefer to spend our surplus wealth on protecting ourselves against. By moving or rebuilding and as necessary to live in harmony with the changes the planet imposes on us.

    The so called anthropogenic climate change problem is in simple fact an unprovable either way soft “science” belief. These have always been a perfect vehicle to exploit as an act of faith, much like an ancient reigion, deceiving the faithful for power and profit with pretended knowledge, and attacking any “heretic” who points out that the inconvenient truth that things may be otherwise, and what is being done in the name of the problem is making it expensively worse, to profit the religion’s insiders, at the people’s expense. Build more pyramids, pay more tributes, make more sacrifices, line up your first born children, the rains will come. I bet the Moche priests had stashed away a tidy retirement hoard in Bolivia well before they made a run for it across the Peruvian Desert..

    Point? Climate science is and can never be a real science, so incapable of proof or disproof, ever. Get over it. At best a rather undersubscribed numerical model that may successfully exptrapolate a few years temperatures but most iikely won’t, without “correcting” the data and model. Not worth the hot airand bytes..

    Better to read the proven science in “Sustainable Energy – Without Hot Air”?

    Your science may vary. I prefer stuff I or an independent other could repeat myself, given the resources.

    Climate change, as advertised by its climationists as being from scientifically proven causes we can modify according to climate change laws, is total BS (Bad Science). What is done it in its name is making it expensively worse on the measured emissions facts, mostly, while compromisising the energy infrastructure we need to maximise wealth to combat the real effects of a changing climate, for whatever unknowable reason.

    Bye again,

    “It was a boring conversation anyway”

  33. In efforts to publish, the results of experimental tests are reported as long as they meet the 5% significance level. Even if the theory is worthless, one out of 20 tests will meet that criterion. By lying to themselves, and throwing out “flawed” data points, experimenters should be able to increase the number of tests meeting the criterion to around 5 out of 20 ..I’ve fallen into that trap myself years ago, when trying to devise methods to handicap horse races.
    There should be an effort to reproduce tests before any paper is published. That should cut down those false positives by 05%.

    • “There should be an effort to reproduce tests before any paper is published. That should cut down those false positives by 05%.”

      0,5 %? I think the 80-90 percent figure is more appropriate when it comes to climate science. For the sake of fun, only a consensus of 7 per cent would remain. This is the best way to tackle the problem of climate science without having to put these fraudsters in detention. On the other hand in science, the problem settles by itself. Those who have productive work in the daily struggle of companies to market shares of their products do not have time to publish peer-reviewed works in journals. They are measured according to the quality of their products and services. The small portion, which still has time for such kinky slits, is negligible. They belong to the 7 percent consensus community.

      • The excuse of “climate scientists”, which is why their work is not reproducible, and secondly, they often show results that do not correspond to the observations, is a stereotypical one. In this way, the climate sciences are so exaggerated, so that this would be the summit of all science, practically the mother of the sciences. An abundance of data to be processed in no science would be responsible, and these data would still fluctuate in an unpredictable manner. New insights would turn up daily, which made the old work no longer reproducibly reproducible. This is the old hare and hedgehog principle. A pseudoscientific excuse for misguided scientific work. If a NASA space engineer were to use the same excuses for a mission to Pluto or a mission like “Osiris-Rex,” the approach to an asteroid, for a bust, the responsibility would be the senseless consumption of several billion dollars (or euros- my currency, that is pretty much the same at the time.). Simply a sick science.

      • sorry typo for 95%. You’ll still reach the 5% significance level for 5% of tests by random chance if the hypothesis is worthless.

    • Oh, the number in my first sentence must be 5 per Cent. But this does not change my statement.

  34. So I’m thinking, “what the heck??”. All these jammering and whinging comments about “science”, everyone thinking they are talking about the same thing. When, if one actually tries to parse the comments, one realizes that everyone is using a different personal meaning of the word. Apples and oranges, bananas and paint, everyone is just tossing mud.
    It would be the same if you were using the word “set”. It is possible to have 464 discussions, each using the word “set”, ALL different.
    Tower of Bable.
    Please, please, use the proper word in the proper place Just because a common usage of the word “science” is to mean “knowledge base”, using that definition in a discussion does not serve well.
    Please be consistent, if you wish to be understood in your rants.

  35. I am a physicist who regularly submits to Nature. Scientists submit important papers here because it has the highest “impact factor”, so it is good for advancing your career.

    All my colleagues know that Nature has the reputation of accepting papers for publication primarily because of impact, and not necessarily being reproducible. We even joke that papers published in Physical Review Letters are more likely to be correct. I am happy to hear they will tighten their acceptance criteria.

  36. There is a reason drug trials are double blind. Even unwittingly an experiment can be biased. That makes reproducibility indispensable, at least for sciences influencing public policy.

  37. Every hypothesis begins in the philosophical logical domain. With deduction and reproduction it may either move or intersect with the scientific logical domain characterized by accuracy that is inversely proportional to the product of time and space offsets from an observation frame. It may also move or intersect with the two other logical domains: fantasy and faith.

  38. They have discovered that correlation does not necessarily imply causation. The use of inference, liberal assumptions/assertions, and abuse of correlation has created a post-normal science where theoretical scientists (e.g. climate science) have progressed to conflation of logical domains.

    • Introductory statistics courses teach that correlation does not prove causation – an often repeated fact. However, it is seldom stated that the corollary- “Lack of correlation disproves causation” is also true.

      • I don’t think that it does. Lack of correlation implies but does not establish independence.

  39. Some of the causes of problems in medicine:
    1) It has been found that the HeLa cancer cell line has been found to have infected other cell lines used in research, as have certain others. With aggressive cancers, such lab contamination is a big problem. If you think you are testing a cancer drug on colon cancer but it is really skin cancer, you will have failure when you get to clinical trials. There are tests for contamination but little incentive to make sure the problem is not there.
    2) It has been found that white lab rates are overly sensitive to environmental stressors due to their inbreeding. In addition, the overly clean environment in the lab has made their gut flora abnormal, which affects their immune system. This has been known for some time but these rats keep getting used.
    3) The problem of data dredging has not been adequately dealt with. If you test hundreds of hypotheses about diet and cancer, diet and heart disease, you will get false positives. So we get headlines that bran muffins will save your life (remember that one?). This problem is also big in the social sciences like psychology.
    These are just the most obvious big problems that are allowed to continue because everyone is desperate to publish.

  40. Try to get something published that reproduces the results of another paper. Try to get funding to determine if the results of an already published paper is reproducible.

    Until funders are willing to fund, and publishers willing to publish papers reproducing the results of other papers, reproducibility will be a non factor.

  41. Irreproducibility makes for flawed science but the opposite syndrome, argument by way of ineptitude, is equally as bad.

    The case of Pons and Fleischmann is probably the best example of such. Their experiment had a controversial result which the textbooks said could not happen. Getting this result called for some painstaking and meticulous work. If the work was not done correctly then the result would not be seen.

    The initial attempts to replicate their work failed, and they were called frauds on the basis of that failure.

    However, it was later shown that Pons and Fleischmann’s results are indeed reproducible, and that the problem with the negative results was a lack of competence by the first group of reviewers.

    It should never be assumed that because one guy says, “I cannot reproduce this experiment therefore it must be impossible” that this is true. It could be simply a case of not having the necessary skill to achieve the original result. Along with a large helping of confirmation bias, of course.

  42. “Sylvia Earle preaches science to students” was a headline on the front page of the Corpus Christi Caller March 3 edition. Her talk was part of the Texas A & M University-Corpus Christi (one of the takeovers by a major older school) Distinguished Speaker Series Student Forum. Previous speakers included a NASA director, a theoretical physicist and Bill Nye. Earle is a celebrity oceanographer/activist connected to the school, but mostly with National Geographic. She was quoted as saying “I just look at the evidence.” She is involved in “maintaining diversity and to curb accelerating climate change.”

    I would argue that the larger problem is elevating hypotheticals, the evident danger sooner in engineering. Hypotheticals are the essence of science, but sometimes sit around for decades or longer before the interest and technology comes along to investigate.

  43. The big science breakthroughs are over. Has been for decades. What remains is the ‘science’ of nearly imperceptible ‘correlations’. A lot of science is just statistics now.

  44. With regards to BBT, the classic Origin of the Elements by Alpher, Bethe and Gamov is still valid.

  45. the big bang is NOT a theory is is a rather bizarre hypothesis that is falsified by what we observe………to claim all the matter in the universe was in one tiny location is LUNACY.

  46. A good start would be to uphold the demand to save data and methods together with the paper.
    Without that it is just an essay.
    It could not be a problem now, when you have plenty of space to save anything. The problem might be that it could show sloppy practice.

  47. I saw the Carlson – Nye debate. Carlson put Nye in a corner by asking him what percentage of global warming is caused by human activity. Carlson pointed out that if the human contribution to global warming is “settled science”, then Nye should be able to cite what percentage humans contribute. Nye squirmed and evaded, but of course he couldn’t answer the question. As is often the case when Carlson takes on liberals, Nye looked like a fool.

  48. Bill Nye couldn’t “slam ” whipped cream . He dodged Tucker questions and when pressed on the question of what percentage of climate change was caused by humans he finally threw out !00 % . Reassuring to know without us pesky humans there would be no climate change . Who knew ? Dinosaurs ?
    One can only surmise that the climate Armageddon industry cooked up their own definition of “climate change ” in secret handshake meeting when the scary global warming campaign started to unravel and they needed a rebrand . They play this little slight of hand trick demonizing people who acknowledge the climate changes but just not their definition of it .
    Nye’s stubbled as he struggled to spit out the 100%
    figure someone looked to be feeding him just off camera .
    Knowing if he said any other percentage they would have to prove it and they can’t .
    Yep that’s settled science all right .
    Funny how no one in the so called ” settled science ” clique ever talks about their self made definition of “climate change” which is vastly different than the general publics .
    The human contribution of CO2 is a fraction of 1% of the earths atmosphere and is a
    chump change factor with mostly positive benefits to the climate while acting as highly desirable plant food helping to green the planet .

    Scary climate promoters rely on a largely scientifically illiterate media that don’t even know what questions to ask. An exception is Tucker who won’t let climate illusionist’s pull the wool over on viewers .
    It is good to see more and more actual scientists in the climate science field and others
    no longer prepared to watch unscientifically qualified promoters speaking about their politically polluted field .
    Scary global warming will magically disappear when the money dries up and comedians stick to comedy .

  49. There’s a lot of wording around scientific practice that talks of experiments and their results.

    Most of the natural sciences don’t involve experiments so much as making observations and collecting data, and trying to draw conclusions. Not the same thing at all. This blog and its commenters do a lot of poking fun at conclusions that go beyond anything implied by the data collected, and rightly so.

    It’s not enough just to go out and make observations any more; now you have to dramatize it by drawing unwarranted conclusions. Not the same thing as reproducibility, but a symptom of the same malaise.


    Off-topic but the latest “oldest fossils” are getting perilously close to the 4.5 billion year age of the Earth. Makes you wonder if the seeds of life weren’t there before the Earth formed. Not my idea, but I’m starting to think it might have merit.

  50. If cooking was done the same way “science” is today we would all starve, or die from food poisoning. Swamp draining, not just a thing for DC.

  51. Pertinent to this issue is the tract of IPCC AR4 in which falsifiability is said to be outmoded and replaced by pal review oops “peer review.” In the corrupt political order that is promoted by the United Nations it is not observational data gathered in in repeated trials but rather are the reviewers of manuscripts submitted to the journals for publication that determine whether the claims of models are acceptable.

Comments are closed.