Slashdot Mirror
Scientist Says Most Scientific Papers Are Wrong
An anonymous reader writes "According to epidemiologist John Ioannidis, the majority of published scientific papers are wrong. If Ioannidis's own paper is right, a randomly chosen scientific paper has less than a 50% chance of being true. He also says that many papers may only be accurate measures of the prevailing bias among scientists. However, a senior editor of a scientific journal says that scientists are already aware of this: 'When I read the literature, I'm not reading it to find proof like a textbook. I'm reading to get ideas. So even if something is wrong with the paper, if they have the kernel of a novel idea, that's something to think about.'"
Great... watch the Creationist/Intelligent Design kooks run with this.
Trolling is a art,
Wow! Science can be wrong.
That is how the system works.
But just because these two scientists were wrong about the precise mechanics of evolution doesn't mean that they were wrong about how the data should be interpreted. The data shows that life has progressed to meet the demands of its environment. Survival of the fittest is correct, but there is no straight-line progression of lifeforms leading one from another as was supposed when these authors first penned their ideas.
Scientific ideas may come and go, but the data set just gets larger. That is why this guy can claim the others are wrong: he has a better data set.
"Rocky Rococo, at your cervix!"
Their is a 50% chance that that's not true.
Scientist Says Most Scientific Papers Are Wrong
I can't believe it!
"I cannot think of any need in childhood as strong as the need for a father's protection." -- Sigmund Freud
So if 50% of published papers are wrong, couldn't this be wrong? But if this is wrong, then maybe 50% of papers aren't wrong. Leading to this being right and creating a horrible paradox...
"Science" is NOT the same as "fact" or "truth". It is a METHOD -- a PROCEDURE one follows in an attempt explain some event or phenomenon. It should hardly be surprising that "Scientific" papers are mostly wrong. There may be only one "right" or "correct" theory for a given phenomenon -- but there are countless wrong ones.
Not too surprising.
I wish people would be a little more weary of automatically believing everything they read in a scientific paper, or worse a crap article from a journalist who doesn't even understand the paper in the first place.
autopr0n is like, down and stuff.
Wow, I thought Study Shows One Third of All Studies Are Nonsense is bad enough, who knows scientific papers are worse!
I patiently await the next article: "Research Shows Three-Quarters of All Researches Are Bullshit".
Rock that crushes, Paper & Scissors that don't matter.
83.72% of statistics are made up.
So even if something is wrong with the paper, if they have the kernel of a novel idea, that's something to think about.
:-)
Whether anything anyone says is right or wrong, it's a matter of opinion first and foremost. Our biology does not provide us with the ability to view things from a topological perspecitve that extends beyond our fixed notion of an asseblage point; our peepers, the mind's eye, whatever you will call it. Therefore anything that anyone says is simply an opinion. It could be an opinion based on data, or facts long believed, or based on stimulus, but it doesn't prove truth at all and it never will.
The day our species extends beyond the fabric of time, space and matter, is the day we will know something closer to the truth. Not beforehand.
Oh we might get lucky! Let's face it... what are the odds that this planet would exist and support life? I think we won the lottery, really.
I can see where Ionnidis is coming from because I'm wrong half the time. I might be wrong about this comment too, but I'm going with my gut feeling!
It is however, the duty of humanity, to reach. In reaching we may fall, or fail, but it is the effort that rewards and replenishes, not the result.
That's why I like Slashdot... it's the ideas and the thought provoking commentary. Don't mod this funny -- I'm being serious.
Story A is posted, 200 people weigh in on everything from the topic of the story, to the troll of the month. The moderation puts most of the good ideas provided in the first five minutes up to a very high score, and the rest of the comments are buried.
So I don't look at the scores as much as I look at the titles. If the title is thought provoking, I read the comment and reflect on someone else's point of view, which is usually highly debatable.
The dangers of knowledge trigger emotional distress in human beings.
Gee, i didn't know most of "IEEE transactions on Image Processing", "Journal of Algorithms" or "IEEE Transactions on Pattern Analysis and Machine Intelligence" were probably wrong.
Please be more specific next time. Thank you.
What do you expect?
That's probably because most scientific papers are put out to be peer reviewed. Most of them most likely do not pass that test, or at the least, are redundant.
// file: mice.h
#include "frickin_lasers.h"
Does it also mean that unpublished scientific papers are right 50% of the time?
"It's the height of ridiculousness to say for those 9 lines you get hundreds of millions."
I thought 33% of the papers were wrong.... Then or this has 33% chance of being wrong, or the other has 50% chance of being wrong....
:-)
Can someone calculate, based on above, a better estimate on the chance of some paper being wrong?
note: don't publish this calculation on a paper, otherwise it will be subject to these probabilities and we'll have to recalculate all over again....
So Long and Thanks for All the Fish!
I worked nearly two years on a project involving a cell cycle receptor. Had some manuscripts ready to go, even had a drug company interested in the potential. Turned out we were wrong. We had based a lot of the research on some commercial antibodies, that despite all appearances, were not working correctly. (Sh)It happens.
Damn good thing truth is constantly changing. Now we can bask in the comfort that those who we've setup as gods are wrong about half the time. So there's a 50% chance Creationism is right? And since they reach each others writings to get ideas instead of facts, perhaps all those people suffering in the wake of Katrina simply had differing ideas about where it was landing. Hey, they had a 50% chance of being right.
The lesson to take away here is that you should listen to a study because it makes sense and presents its argument well, not because it is "a study".
But if you didn't realize this already, then I think you have a problem.
I don't know if I should believe him or not, couldn't his paper be wrong too? What if textbooks were wrong too? Movies? What if the dinosaurs didn't even exist? I can't handle it anymore!
I have found this to be sadly true. My coworkers are college professors who often publish papers on social trends using large datasets obtained from government records. I am frequently pointing out errors in their analysis (mainly that they simply don't look at their data, such as just because Jane Doe and John Doe have the SSN they are assumed to be the same person) but I'm generally ignored or told to fix it myself though that isn't my job. They are more interested in getting something published and don't want to have to retract something.
Science seeks the truth but never claims it. If someone claims something is absoutly true they are selling something.
If loving Science is wrong, I don't wanna be right.
Selfish link to Linux/BSD Gangster. We need more OG's cracka
What I do see as harmful is the attitude towards bad papers. To many academics try to accumulate more and more published papers the same way that slash-dotters try to build up karma. I understand that having papers published can reflect well on someone, but we need more accountability. Journals need to create a more strict system for reviewing papers that are to be published to weed out more of the crap plain and simple. If the evidence does not reflect the claims throw it away. If the research was conducted on a population that was too small or specific for a grand generalized claim about the topic as a whole, throw it out.
I understand that you will always have people just trying to throw their names around, but this needs to be looked at from the grander perspective.
"When I read the literature, I'm not reading it to find proof like a textbook.
Sure there are probably many scientists that think of it this way. But the problem is that bad research (or a bad paper) rarely dies after being published. They are often cited as evidence for years to come in other papers until enough evidence to the contrary comes out to raise questions. Plus, you have crazy professors giving this bad research for their classes to read, and often they don't explain to their classes where research is possibly flawed--so we find ourselves training generations of new scientific minds that run around spouting out bad research. I understand that we all need to take research with a grain of salt when we read it, however bad scientists trying to become famous with their bad ideas or bad papers can be very detrimental to any field.
If the researcher is Derek Smart, then there is a chance the research ITSELF isn't just wrong,/i>, it DOESN'T EXIST!
http://en.wikipedia.org/wiki/Derek_Smart
How certain are we that this guy isn't part of the other 50%?
... that this paper falls in the wrong 50%.
You "reward" scientists for writing papers like mad, expect all their work to be significant and their hypothesis to be proven right everytime so the 'research-dollars' 'feel' well expended.
Most editors would not publish a paper titled "I tried something reasonable but IT DID NOT WORL AT ALL!".
Comment removed based on user account deletion
But seriously, reviewers are biased and sloppy, as are the editors. The fact that reviews are blind means that they are also unaccountable, which fosters even more bias.
Journals take months or years to respond to a submision, and often as not they respond with a rejection so the submitter has to give up or start the whole process over with another journal. There are so many scandals that one could quote. The whole process seems more designed to support the status quo than to promote knowledge.
I have discussed this with many people in academia and they react not with logic, but with horror that I would dare to question a system that they view almost mystical reverence.
Test 1 2 3 4
73% of all stats are made up on the spot...
22% of all people know that.
If a submitted paper is scientifically unsound, it should be rejected.
If a scientific paper is useless to the readership, that publication should reject it and recommend a different journal.
If a paper is wrong and the reviewers KNOW IT then they should send it back for corrections.
If it's WRONG but the reviewers don't or more typically can't know it because it is novel, then publish it. The rightness or wrongness will be sorted out soon enough.
Ever heard of Isaac Newton? Turns out his theories were incomplete in some very fundamental ways, but his theories regarding the motion of objects were the best approximations we had for hundreds of years and are still very useful for macroscopic objects traveling way below c.
Knowledge is how to play a game, intelligence is how to win, wisdom is knowing what game to play.
Research shows 70% of slashdot editors are easily amused by bullshit.
"Facts are meaningless. You could use facts to prove anything that's even remotely true." This is clearly the exception that proves the rule.
Q: What did the comedian say to the crowd?
A: If I knew, this joke would be funny.
Newton's original papers on physics are all wrong. So what?
They've been replaced by something else. Sure, they're generally true, enough to be taught in physics classes, but all the specifics on gravitation etc. are incorrect.
They're being replaced with: (pick your theory) quantum gravity, string theory, quantum mechanics and more things I don't know.
But so what? Science, by its nature, is always being improved upon. Any time you correct someone else's theory, you could say their theory is now wrong.
Well, maybe this description is even wrong or inapplicable, considering I didn't read TFA =)
rofl
If you get nervous, just remember that there are a few billion other people who don't really give a damn.
Clearly he should write a paper about it
---
The whole point about science is that you build on the work of others. Newton coined the phrase "standing on the shoulders of giants" and his work gives a good example. He formulated a theory of gravity which was improved upon by Einstein in his theory of relativity. That, in turn, is now being partially questioned (I believe, but I may be wrong...). Therefore it's not overly surprising that many scientific papers are incorrect. People can correct them later, and often do.
I know of a whole company based on a bad paper. Some type of "fast blood analyzer". After a number of bad pre-production starts, yelling matches between software and hardware people, firings, suings, quitings, it was finally determined that whole premmis (from a founder's scientific paper) was false.
-- www.globaltics.net
Political discussion for a new world
Most researchers only read papers as a base for their own research. Speak to almost any established scientist and they have usually learnt to not believe everything published the hard way (ie they spent months chasing after an artifact in someone elses data, without checking it first themselves). It would be impractical to do anything about the situation as journals can't ask independent groups to re-check every piece of data for every paper submitted. What does need to change are the journals that are acadamies based (PNAS springs to mind), where it is still sometimes possible to get manuscripts published without a true peer review process (ie you get your acadamy friends to "review" it for you, rather than an independent source).
"If we knew what we were doing, it wouldn't be research"
That's what my supervisor used to say to me when I got depressed about lack of progress.
Just because 50% of papers are incorrect has no real baring on the integrity science. Because its not the publishing of papers that make a theory correct.In Science for a theory to be accepted as being 'fact' it must be testable make predictions and be falsifiable. Ioannidis says "We should accept that most research findings will be refuted. Some will be replicated and validated. The replication process is more important than the first discovery," Ioannidis says. Ehh Yeah Mr Iannidis you need to take a lesson in the Philosophy of Science thats the whole reason of Science you publish a paper its gets checked by its peers then the if its hold up to further test and prediction it becomes stonger till it eventually gets universal approval! Duh The real problem is when a paper gets published the press get hold of it, sensationalise it before its ever gone through the proper scientific checks i've just mentioned.
The issue is that statistical significance is typically reported without regard for correction for the testing of multiple hypotheses. In fact, in areas like "whether a particular gene influences a particular disease", this is a rampant problem, and Ioannidis is hardly the first to identify it. (In academic studies; clinical studies that are being submitted for regulatory approval have to prespecify their hypotheses and don't have nearly the same problems.)
But this has nothing to do with the overwhelming majority of "scientific papers". I doubt if one chemistry paper in a thousand is affected.
What I'm listening to now on Pandora...
According to epidemiologist John Ioannidis, the majority of published articles on Slashdot are dupes. If Ioannidis's own paper is right, a randomly chosen story has less than a 50% chance of being original.
This is the fucking unfunniest bunch of comments that I've ever read on Slashdot. Geez!
the irony!
CBF reading the actual paper for now, but the New Scientist writeup is pretty stupid for such a reputable publication. For example, they said:
"If you test 20 false hypotheses, one of them is likely to show up as true, on average."
Well, duh. Ever heard of a Bonferroni correction? Morons.
A lecturer at my old university told me that around 90% of papers get written, then put into a drawer somewhere. And nobody reads them again. I wonder what proportion of papers that are read are 'wrong'?
Anyone who has followed the dinosaur debates over the past few years should not be surprised by this statement. A few examples:
A feeling of having made the same mistake before: Deja Foobar
...once people KNEW the earth was the center of the universe. ...once people KNEW the earth was flat. ...once people KNEW we were the only life on ear...oh sorry, got carried away with the MIB reference.
Anyway, the entire point of science is to learn the 'truth' of the universe. There is no endpoint. You can't 'prove' anything with any finality because there's always a chance for an exception, a chance, etc.
Hell, they just made something go faster then light from what I've read. That invalidates quite a few papers.
There's a difference, IMHO, between flat out WRONG and an old idea being proven inferior because of new facts, etc. Which is what science is. If this wasn't true, we would already know everything. The constant replacement of what we know with updated, new, and more thoroughly examined ideas/theories is the entire point.
So while it may be said that statistical studies may not always be as conclusive as they're made out to be, generalising 50% of published scientific literature to be wrong is simply Troll.
I don't want to read
At least, in the field of the life sciences, a paper that simply verifies a claimed result is considered to be a quality paper that is worthy of publication. In the field of computer architecture, virtually no one ever verifies another researcher's published results because verification is considered to be a dull waste of time. The consequence is that computer conferences like the International Symposium on Computer Architecture (ISCA) is fraught with extreme claims based on results from simulators that are considered "reliable" simply because they do not produce a core dump.
Worse, in addition to the non-incentive of verification, there is also the practical problem of using the right simulator. During the 1990s, the bulk of the wild claims in ISCA papers were based on the MIPS instruction set architecture (ISA). Most universities had SPARC machines and were ill-equipped to run the MIPS-based simulators. Alas, verification is virtually impossible.
John Ioannidis, an epidemiologist at the University of Ioannina School of Medicine in Greece, says that small sample sizes, poor study design, researcher bias, and selective reporting and other problems combine to make most research findings false. But even large, well-designed studies are not always right, meaning that scientists and the public have to be wary of reported findings.
... you have to be careful.
OK, I'm going to go through these one by one.
First, small sample size is a problem. That's why you have error bars on your graphs - in fact, if you don't see the error bars, check the tables to see if the t size is big enough - many studies start with thousands of inputs to get only a handful of outputs - in biochemistry, you can have more than 10,000 PCRs of something made, only to result in 10-40 final structures in crystallography at the other end of the pipeline.
The study we're on is unusual in that it actually has sufficient numbers that the t sizes are big enough to ask many questions - but most have such small numbers that they could easily be wrong.
2. Poor study design - again, how you ask the question is important, as well as the conditions - so this may be true. I always check the holes in the logic as well as the basic logic - because those holes can lead to incorrect conclusions - and many popularized science articles don't bother checking for the holes in the logic. They do a quick summary saying "breast cancer is caused by too much salt in the diet" when the study really said "there is a high correlation among middle-aged women having first onset breast cancer if their diets are in the top range of salt intake" - but that could also mean they live in conditions where the high salt intake could be due to the other things in their environment that caused the breast cancer in the first place.
For example, you could say Romans got lead poisoning because they lived in cities, when it was actually the use of lead in their pipes, not the living in cities - although we don't know, as perhaps cities had lead particulates in food from airborne fallout from factories or burning certain things in their candles
3. Researcher bias - ok. Not going to argue that.
4. Selective reporting - see 2 for how this occurs.
But that doesn't mean a good high-quality peer reviewed scientific paper in a respected and well-juried paper is "inaccurate". There are a lot of journals out there, and different standards and quality levels.
-- Tigger warning: This post may contain tiggers! --
Scientific papers often include a great deal of data and analysis. Some of this data can be somewhat inaccurate, much of the analysis can also easily be incorrect. How far off does something have to be before it is "wrong"? How much of a paper has to be "wrong" for the paper itself to be declared "wrong"? I think a better way to look at it is that in most papers, there is some wrong and some right.
Ceci n'est pas une sig.
most papers on God are wrong.
Game: Player 'Donald J Trump' now has AI skill level 'experimental'.
The problem is that it's hard to get funding. It's a lot harder if your last 5 papers were in the Journal of Negative Results because the experiments didn't support the theories. Scientists know how to design good studies, and they know how to do thorough work. But, it's not always in their best interest. If some of the data point one way and some point another, it's a lot easier to create more opportunity for further grant monies (for everyone in the field) by being selective about what's reported. They're declaring victory, in other words, albeit in a subtler way.
"All models are wrong; some are useful" -George Box
I'd say it's more like 100% of scientific papers are wrong, it's merely a question of the limitations of the model.
Depends upon whom the peer is.
or didn't you mean that kind of peer...
A feeling of having made the same mistake before: Deja Foobar
It must be true that most scientists are wrong: If most scientists say that most papers are wrong, and the papers are really right, then most scientists are wrong in saying that most papers are wrong. Otherwise, if most papers are wrong, then most scientists are wrong for haven written those papers.
Ceci n'est pas une signature.
If 50% are wrong, then 50% are right. So if I write a scientific paper, the chance of it being right is 1/2. And if I write the same paper, say, 8 times, the chance of it being right at least once is 255/256.
I think I'll write that paper on statistics.
Indiana Jones:
Archaeology is the search for fact, not truth. If it's truth you're looking for, Dr. Tyree's philosophy class is right down the hall.
"Lawyers are for sucks."
- Doug McKenzie
Science is a process by which we replace a less accurate description of the world with a more accurate one.
-- Sig down
It took them 100 years but they spinned creationism into ID, in 100 years they'll come around. Evolution has always been evolution, their side twists and spins. The truth will always be right.
2 years and no mod points. Join reddit. Because openness is good.
Peer Review.
Slightly offtopic; but don't people know that using the phrase "as of now" is wothless in a reference article that may exist for a long, long time?
What if this paper turns out to be wrong!
What could it mean?! Ahhhhhhhh!
And 50% of Zonk's posts are dupes.
(Yes, it's a dupe. Both stories are about the same study by John P. A. Ioannidis, MD.)
-Peter
set of all sets...
Lars T.
To the guy who modded me down from perfect to terrible Karma - Apple haters still suck
Does this mean Slashdot is a scientific journal?
this is why math is the only true science left...you can see for yourself the truth or falsehood of someone's paper by looking at the proof that is offered IN the paper! imagine that!
I would predict much different results if the question had not been "what percentage of science papers are wrong" but "what percentage of your own science papers are wrong".
A better survey would have discounted ego by asking both questions.
To spell it out: The possibility is that each scientist believes axiomatically "I am right and others are wrong". Because the majority of papers which any scientist reads are written by other scientists, each scientist judges most papers to be wrong. The survey results confound the egos of scientists with the accuracy of their papers. The difference between the average ratings of scientists judging their own papers vs. those of others is a measure of their egos.
Ceci n'est pas une signature.
In my experience it is not quite as bad in my field (CS), but I have made some bad experiences here too, including papers accepted at respected conferences that had completely wrong measurements and these were the basis for everything else.
The true problem starts when reviewers and superiours (like PhD advisers) are not competent enough or unwilling to see that some seemingly great results are more likely the result of extreme pressure and (at best) educated guesses. Also quite a lot uninteresting results get published because of politiking. Almost as bad IMO.
Most ACs are not even worth the keystrokes to insult them. Be generically insulted by this and ignored otherwise.
What you say is very true. I am working towards my Ph.D. right now in Physics, and I encounter this every day.
I think both you and the poster hit on something very important here. First is that we (as people who are reading the research papers) are not looking for proof, etc. of something. When I grab the latest paper on a topic I work on, I am not going to read it and say, "Oh. They found x which contradicts with what I am seeing. They must be right." Instead, I am looking at their models, results, and the like to see what HAS been done, what the outcome was, and if they have any problems. That way, I can address these issues in papers I write and talks I give. It also gives me something to compare my results to.
Your cousin is facing what I am facing, as well. I think people in the community understand perfectly well where she is -- that she has found interesting results thus far and needs to work more on it to get deeper understanding/whatever. To the outside world though... "it's just a theory" or "it's just preliminary" (phrases people love to throw around) drowns out the important stuff she's doing.
The second thing, that you specifically hit on, is that we need to eliminate options when we're working in an area. The pressure to produce sexy, nightly-news-ready results keeps us from doing that. Of course, I'm biased. Hah.
This doesn't mean all of this work has been pointless. People do studies, report their results (which are sometimes wrong)... but we know what we're doing, and we know what to take with a grain of salt and what not do.
Mike.
Mmmm......sacrelicious.
The paper stating that ~50% of scientific papers are false is published in the Public Library of Science (PLOS) Medicine. The paper only examined medical studies and not scientific papers on physics, chemistry, engineering science, (and mathematics).
While molecular biology papers can be prone to statistically insignicant, but factually stated conclusions, the biggest culprits are clinical studies and 'large-scale' analyses of data.
Good experiments are constructed to give a 'yes' or 'no' answer based on the presence or absence of evidence. The zeal of high-throughput studies and analysis have put more pressure on good statistical analysis. Unfortunately, statistical analysis requires math...which sometimes eludes doctors and biologists. Hence, the problem of missuing statistics and stating inadequately supported conclusions.
-Howard
Favorite
Science progresses when well thought out hypothothies based on a good data are replaced by more inciteful reasoning based on more complete data. Lamarck wasn't guilty of faulty reasoning. He just didn't have a complete enough data set.
But the article at hand, isn't talking about that kind of "wrong". He is talking about conclusions that can not be supported by the data presented. Either the reasoning is faulty or the data collection methods are so faulty that no meaningful conclusions can be drawn.
When a theory is proven wrong in the scientific sense, it is a good thing. We learn something new and that be the basis for further developments. But if a theory is proven "wrong" in the mechanical sense, we have no new insights, just a relief from further time wasting.
> Yes, but unlike religious dogma, scientific theories are meant to be falsifiable.
That's unfair! they can falsify ours and we must say okay, but we can not take down their bullshit because to them and their listeners they are always right?
Does this mean to the average american extremist, that we can only lose??
Any sufficiently advanced intelligence is indistinguishable from stupidity.
Wait.... scientists arent biased! They are objective, they want to find out the truth right? RIGHT?!?!?
I don't get it.
He is basically saying that *his* paper (if chosen randomly) has a 50% chance of being incorrect, which, if so, means it is true...
The secret of success is honesty and fair dealing. If you can fake those, you've got it made. (Marx)
All your answers are in Teh Bile-Balllllllll!
Bile-ball == interesting (and somewhat disturbing) visual
Granted, bile-balls could be thrown around rather like invective - and to the same effect - so I guess it makes sense regardless of the Freudian-slip-iness of phrase.
I want to drag this out as long as possible. Bring me my protractor.
Science is not about finding the truth. It has nothing to do with the truth; people who look to it for truth misunderstand it. Science, and the scientific method, are based on one thing: reproducible effect. I do X, Y, and Z, and T results. If this can be confirmed, reproduced independently, you might have something scientifically useful.
Notice what this does not say: X, Y, and Z are "true"; Z is "true"; X, Y, and Z cause T. Nor does it state the meaning of X, Y, Z, or T. Nor does it say why, in the presence of X, Y, and Z, T occurs. These are irrelevant. The only thing science does, the only thing it is capable of, is one thing: testing if, in the presence of X, Y, and Z, we repeatedly get T. For most things, that's all that matters. This is the scientific method.
Thus it is that science is, quite literally, magic. Look over most fictional magic systems. We have things like "if we say this spell, this thing happens." "If we write these symbols, this thing happens." "If I visualize this thing in my mind, this thing happens." "If a mix a pinch of this and a hair of that, this thing happens." Because it's reproducible, it's useful. The mechanic does not matter: only reproducible effect matters. If waving ones hands and saying a phrase were to be followed consistently by a minor explosion, it would be just as scientific as mixing two chemicals to produce the same effect.
It doesn't matter why. Theories get revised consistently to fit the facts, to document reproducible effects. If phlostigen and ether were accurate and useful models, the fact they have been discarded for more useful models does not matter: science isn't about truth. It is about reproducible effects.
This is why not clinging to pet theories (yes, this includes everyone's favorite: natural evolution) is important: the theories do not matter. One should never fit facts to a theory. One should create a theory to fit the facts.
This is what makes science useful.
Don't think of it as a flame---it's more like an argument that does 3d6 fire damage
Of course, google is only a very statistical method
:)
Should read:
"Of course, google is only a very gross statistical method".
And yes, errata is a fundamental part of Scientific research, too!
Ya and 47% of statistics are made up on the fly.
...that all depends on how you look at it.
I suspect many papers (e.g. SIGGRAPH ones which I read regularly, and are thus technically computer science) don't have an objective "right" or "wrong", but do have a conclusion. Often, this conclusion is subjective (something which looks or performs "good enough", or is a workable approximation, etc).
So yeah, maybe half the papers are "wrong" in some sense, but that doesn't mean they aren't useful, or indeed that they aren't known to be "wrong" by the people who create them!
Game dev and music blog
When the people with the money (private or public!) understand that increased knowledge can't always deliver immediate "answers," we will get fewer "wrong answers"--not necessarily because the research will be better, but because scientists will be more willing to say, "I explored this area and didn't find anything."
Will we get "fewer wrong answers" or just "fewer answers"? I don't know that removing the pressure will be any better.
How many "researchers" will just sit around waiting for the next paycheck if they don't feel any pressure to actually provide answers?
How many "researchers" will just pretend to work and say "sorry, I didn't find anything interesting yet... by the way, I need more funds, thanks you."
How I'm supposed to trust a paper that states that any (which includes this one) random picked papers have less than 50% chance of this paper beign true?? I mean, what makes this one special and 100% true?
It is truly hard to have an original idea- and more and more educated people are competing to have them. With such competition- people are forced to put out papers that may not be up to snuff.
And never mind the professor who already have their higher degrees- they need to continue to publish- regardless of the quality of their work.
As others have said, the scientific method is such that any given theory is put forward with the assumption that someday, somebody is probably going to prove it wrong and come up with a better idea. That's how progress works, and almost any paper that you look at these days will be sitting on the shoulders of hundreds if not thousands of older papers that have been proven at least partially wrong (possibly by the paper that you're looking at).
There's also the recurrent issue that lay commentators continue to evaluate all academic research (whether in the hard sciences or not) on the basis of right and wrong, profit and loss, and current applicability. Research papers should not be evaluated on the basis of right or wrong; it's simply an inappropriate measure. The question that a journal editor should ask is 'Is it scientifically valid?' - by which I mean that the article is methodologically sound, uses reasonable theoretical sources (or sound and convincing argument) and is honest in the scope of the research. There's no real way of knowing whether scientific research is actually right or wrong before it's rigorously tested, reproduced and examined by a large number of people. The same goes for social research, of course.
The cancer research example is perfect - it's obviously valuable research, and information that needs to be published and spread around the medical community. Is it right? Is it wrong? Who cares? It's new information that somebody might find interesting - even if it only ends up as a footnote in the study that finds a cure for those cancers, it's still made a contribution.
You actually have to read the paper to understand it: His reasoning applies to studies using a particular statistical validation as evidence. It's for medical trials, not newtonian physics.
I am Physics researcher and this sort of thing is quite common. Theoretical papers have this problem in particular because usually there is little experimental evidence and the mathematics is so complicated. Even in a peer review situation, many just don't bother over every little detail and one wrong detail can throw an entire calculation off. The point isn't that every little thing that publish must be right but rather the mistake is picked up someone else and eventually some useful truth is discovered. Hawkins, Newton, and even Einstein published something that wasn't correct.
You don't have to be smart to use a Mac, you just have to be smart enough to buy one
But Intelligent design did spark a great open letter - check out the dogma of the Flying Spaghetti Monster at www.venganza.org
Stop intellectual property from infringing on me
The main point of the paper is, essentially, that there are lies, damn lies and statistics. Statistics give a false sense of precision.
If researchers only ran one regression and reported the results of that regression, then standard tests of significance would be fairly accurate.
Instead, researchers run lots of regressions, tweaking variables and data to refine their model and get significant and meaningful results. But this then means that the tests of significance are now bogus. As the summary states, if I run 20 regressions I'll probably get one significant one by chance.
A consequence is that a strong test of a model is whether it can survive the addition of genuinely new data. Importantly, this is not the same as the common test of withholding some of the data set that is available to the researcher and then validating the model by showing that it still works with the new data. The problem is that the researcher knows about this data and will obviously only report a model that works on the full data set. The write up will give the impression that the model was robust to the new data - but the whole process is such that it is practically the same as just finding a model that works on the full data set. The 'validation' is just smoke and mirrors.
A consequence of this is that a lot of papers will be falsified by people getting more data that was unavailable to the original researchers. In essence, despite the appearance of objectivity, the data analysis process and interpretation of statistical significance tests are corrupted by actual practice in paper writing.
Everything that science has to offer is not just based on a logical syllogism, it is also based on scientists having a squeakier clean reputation than preachers.
You can argue falsifibility all you want, but, that does shoot science in a whole. Evolution, for all practical purposes, is not falsifiable. You can't set the initial conditions of the earth, have a control planet, wait 4 billion years, and get the same result according to a mathematical formula.
Global Warming is also practically non-falsifiable. You can't have a control earth and a current earth, in the same orbit, and then alter the CO2 level and see what happens...
Anything that is not provable by immediate experiment is thus speculation at best. It may be informed speculation, it may be internally consistent speculation, but until you demonstrate your system by corporate like repetition, you don't have science.
PS. Computer models are not experiments.
This is my sig.
I'd say that this one falls into the group that are probably wrong.
The problem with the paper is that it is based on the rather trivial point that for a single result, obtained with a statistical P = 0.05, where the prior probability of the hypothesis being correct is low (i.e. most of the time the null hypothesis is true), obtaining that positive result does not give you a very high net probability that the hypothesis is correct.
This is all quite true, but scientists are quite aware of it. This is why it is hard to get a paper published if a major conclusion depends upon a single result with P = 0.05. Generally, the important results in a paper are based on multiple experiments, and often with P much less than 0.05. P = 0.05 is, after all, by convention merely a minimal criterion for an experimental result to be taken seriously--even under the most optimistic circumstances, there is a 1 in 20 chance that the result could have occurred by chance. Even then, few scientists will consider a question entirely settled if the results are all from a single paper, or even a single research group. Scientists are fully aware of the fact that the P value is typically optimistic, since it doesn't take into account systematic error, bias, or artifact, and is often based on unverified assumptions as to the nature of the statistical distribution. So it is misleading to think of a research paper as representing one or more conclusions that are either true or false--it should rather be thought of as a narrative of a series of observations, with the statistics used to provide a general estimate of the level of "noise" present in the system. A published result is simply one more data point which a scientist will take into account in evaluating whether a hypothesis is likely to be correct.
Moreover, the model in which there are a large number of possible conclusions, of which only one can be true (so that the prior probability is small) often doesn't apply. Very many research questions are binary--i.e. there are only two possibilities, true or false, and there is no strong prior reason to prefer one over the other.
Wow: A) if his conclusion is correct then its correctness requires that it be wrong due to the recursive nature of the fact that it is a published scientific paper (.499 to the power of infinity) and B) if his conclusion is incorrect then its incorrctness allows for the possibility of it being either i) correct or ii) incorect. A and B:i are illogical so his paper must be wrong!
It should be very clear, not only from the journal the article is quoted in, but also from the types of studies he is considering, that he is not talking about scientific studies as a whole but more specifically medical studies (although it appears his logic can be extended to the social sciences as well). Many of his arguments, cannot, however, be extended to the pure sciences.
I came here for a good argument
Hate to get into metaphysics, but a scientific hypothesis or theory is only going to be able to predict a very restricted set of things.
People often quote Newton's physics as being "proved wrong" by Einstein's relativity (and those same people often barely understand the limits of relativity with respect to the quantum mechanical world). However Newtonian physics is good enough for most (though not all) space mission planning since it's still quite accurate so long as you don't get near a large gravity well like the sun or travel too fast. So Newtonian physics isn't "wrong" it's just accurate to within a certain margin and useful under less general conditions than previously thought.
That's what these non scientifically trained creationalist types miss. There is no right or wrong theory, even though that's how the popular scientific press reports it. There is only the ability of a theory to predict what can happen (or has happened) based on a set of conditions, and an accuracy under a given set of conditions. Newtonian physics is no more "wrong" than eating salad is. You just can't misuse it by applying it to the wrong set of conditions (don't eat that salad if you're allergic to the ingredients).
These posts express my own personal views, not those of my employer
And also I think he means that 50% of scientific papers are flawed in some major way - not that they are complete junk.
I took structural equations modeling while in graduate school. After taking it, I noticed that I never ran across a psychology paper in a scientific journal that used this particular technique and used it correctly - except for the papers that my teacher used to show us in stats class.
However, that's a very small sample as I dropped out of graduate school shortly after. But, I did encounter many papers that were clearly using LISREL (Structural Equations Modeling) incorrectly. Some of them were reading assignments for other classes where I was asked to comment on the paper and my teachers didn't think much of my comments which were all about how the author had incorrectly used the stats. They were intereseted in the ideas expressed rather than the stats. But the stats were the fun part for me.
Avoid Missing Ball for High Score
Did anyone actually RTFP? It's one of the most spurious pieces of "research" I've ever read. And with a biophysics degree, I have read quite a few. The author actually didn't investigate any actual papers, but he builds a mathematical model out of his own biases, statistical projections, and some back of the envelope computation. Even then, his conclusions are much less stringent than the submitter makes them out to be. He "proves" that under all his assumptions, half the research papers *might* be wrong, but shows not even statistical evidence that they are.
I think PLoS is peer reviewed, but that paper should never have survived peer review. Occasionally, bad papers slip through, even in the so called hard sciences. This one seems to be one of them. Since PLoS Medicine is pretty well respected for an open access publication, lets assume that this was a lark and more on.
But it makes me curious what the fraction of bad papers looks like in an open access publication like PloS versus a traditional journal like, say, Nature, The Lancet, or New England Journal of Medicine. One reservation people have about open access (or author pays) models was that since PLoS gets paid about $1500 from they authors, they might be accepting vanity papers, or don't triage as well as traditional journals. I don't think they are, but if this paper is any indication, PLoS might take a second look at their peer review process.
Yes, remember most "scientific" studies are wrong. This is the fault of what we label as "science": psychology, sociology, and other social "sciences", led by people who attempt to do "statistical" research without a statistics degree.
In the average psychology research project, a biased, non-representative sample set of twenty or so random students (none of which share identical biology, brain chemistry, or even the same upbringing) is typically used to extrapolate to the behaviour of six billion other people.
Meanwhile, a careful statistician spends three months on a simple study to analze the effects of a product line when examining for a single independant variable; and despite care and effort, still sometimes produces flawed results.
If the statistician ever extrapolates his results beyond the production line he studied, it will be with a long laundry list of explictly stated formal assumptions, and these assumptions will need to be independantly verified by plants or companies before his research is considered to apply to them.
How many of these unscientific results were really just due to these social "scientists" rushing in to shoot off their big mouths in places where qualified statisticians fear to tread?
--
AC
The real point is, in the eyes of the common man, science is a brand of information, just like Walmart is a brand for stores or Nike is a brand for shoes, and the brand is taking a beating.
Here's the attitude.
"you want someone to believe human origins from a set of people that told me I would die if I smoked and ate a cheeseburger and I'm still living."
"well now basically you are just making up evolution to fit your story together. Well I can do that too. Can't test it either way, can we..."
This is my sig.
is that 42% of all statistics are completely made up on the spot.
http://www.archive.org/details/ThePowerOfNightmares
"Statistician Says Most Statistics Are Wrong"
According to statistician Dr. Cal Culator...
try going to a random high school in america and telling a science teacher that science can be wrong.
Duh! Look at how widespread evolution has become. All from a new "kernel" of ideas.
humbug!
Science is a process for evaluating empirical knowledge the scientific method,
which is the traditional view of theory and progress in science.
Empirical methods have dominated science until the present day
Recent theories such as quantum mechanics, constructivism, theories such as quantum mechanics
May provide the solidity of empiricism the ability to discover even counter-intuitive scientific laws,
and the ability to rework our theories to accept these laws.
Science ? True : False is it that simple.
scientists publish a scientific paper on how all other scientists but them are wrong, in other news differnt scientists also publish a scientific paper on how all other scientists but them are wrong. in other news the media publish there own papers on how all scietists are right. leading the world dazed and confused and looking for answers. most peopel on earth have joined mormon religion to escape the confusion.
A recent study shows that research causes cancer in laboratory rats.
The point of publishing a paper is not necessarily to publish true facts, it's to put forth ideas that might turn out to be factual. It is up to the peer review of the scientific community to determine the validity of the published papers.
That's the whole reason the system works so well. If some grand moderator were arbitarily deciding what constitutes a "true" paper and preventing it from being published in any journal, we'd have a tyranny instead of a peer-review process.
This would be counter to the spirit of science.
Could it possibly be that he is 50% right so only 50% of the 50% are wrong.
:)
statisticly speaking 100% or stats are wrong.
So what you're saying is there's only a one-third chance that it is wrong?
The paper seesm to focus on the fact that small sample size and limited true statistical significance of study results is the reason.
My argument would be of course most studies have small sample sizes and somewhat tenuous statistical significance. Large studies cost large amounts of money. If there were no small studies that published with small amounts of statistical significance how would larger studies get funded? You have to start small in order to figure out whether it is worthwhile funding a larger study in the first place.
This obviously depends completely on the field of science that you are in. Epidemiology? Please, that hardly even counts as science. You're basing this on a field that you can't even do experiments in! You just wait for an outbreak to occur (fairly rare) and then see what happens and base all of your conclusions on a few isolated incidents. My advice to Dr. Ioannidis is to pick another field where you can do some concrete science.
Science, imho, is the process of explaining the present based on deducing the past, and predicting the future based on observing the present. Truth, with a big 'T', is for dogmatists.
Scientific papers represent hypotheses. They get 'modded up' for reproducibility (for experimental hypotheses or claims), falsifiability (no magic, please), consistency with past accepted theories (as General Relativity subsumed Classical Mechanics), and elegance/symmetry.
The Firesign Theatre told us this in the early '70's - "Everything You Know Is Wrong." They also predicted Slashdot - "I think we're all Bozo's on this bus."
"National Security is the chief cause of national insecurity." - Celine's First Law
When you think about it, that's positively astounding. There are vastly more ways to be wrong than to be right. We've managed to get 50% right answers out of the myriad wrong answers. Pretty impressive!
It would be better still if it was more than 50%, but we can just apply the process repeatedly to push up our confidence (50%, 75%, 87.5%, etc.). A little more attention to statistics would help us raise the base rate above 50%.
"According to epidemiologist John Ioannidis..." Pimple Popper M.D.! Eat nuts - don't eat nuts Drink coffee - don't drink coffee Don't eat carbs - eat carbs Is there any wonder why a MD would think scientific papers are bogus? Maybe he should pick up some physics, chemistry, math, and biology journals.
What?
...
;0)> ), then I'd love to hear it.
So does the question "did the universe have a beginning?" have a good answer? Isn't "good" subjective?
A corollary to your example.
Was it a lion or liger (lion-tiger) that jumped out the bush and ate me? How do you determine the species/genus? What amount of genetic differentiation is required? Is this not subjective?
I may have died (death is quite un-scientific as their are lots of unfalsifiable hypotheses about it, so perhaps I just got re-incarnated?) but you still don't know whether I was eaten by a lion or not. Indeed, perhaps the lion just killed me, perhaps the hyenas ate me. How much of me do the hyenas have to eat to be considered to have eaten me? Is the stuff that the hyenas ate, ie flesh, me?
Oh, and was the lion in the bushes before it jumped out and ate me. Or was this just a sense-data fed into your brain-jar.
So, you don't know if there are bushes or a lion. If there's lion you don't know if it's actually a _lion_ or some other creature that resembles lion. If it jumped out an mauled someone, you don't know whether they died or not, nor if _they_ were eaten by the lion. All of these points are subjective.
Yeah, no _real_ questions ever get obtuse and debatable
I'm sure you think I'm just baiting, but if you have a scientific proof for [starting with the basics] the existence of other minds (than my own of course
Thanks
that in a paper where the author complains about generalizations from small data samples, he himself generalizes his observations on epidemiology papers to all sciences?
the level of religous dogma in some camps.
The only question is, who decides which science is wrong? I doubt very seriously any big money areas will have a published high rate of error. After the high money science the next protected type would be whatever is en vogue for the time.
Scientific integrity took a big dive in the late 80s as special interest groups suddenly realized that marketing, confusion, and intimidation were far better at advancing agenda than honest science.
* Winners compare their achievements to their goals, losers compare theirs to that of others.
When there's no umlauts, you use "oe" for the umlautted o.
Please, for the good of Humanity, vote Obama.
Well actually, ID requires someone with a clue one how life and everyting is created. This someone, lets call it "The Creator" could be defined as god.
It doesn't nessesarily means to be one god, or Christ, Jaweh, Buddah, or just a bunch of aliens but still it requires someone or some group, non human, since they didn't exist at the time.
Therefor god is relevant, if it is proven no one existed to create/design the world and it enhabitants, it will be falsifiable, and thus making it science.
Though there is no known proof of a (group of) gods, on the other hand there is no proof of the non existance of a (group of) gods.
Bit of making it the billion dollar question, ain' it?
Going offtopic: I don't believe in a god, but if it exists or existed it did a moderate job in designing. Nor do I believe in the non existance of a god, it's just something I cannot proof or disproof.
My personal view is that religion is just an invention to deal with bad luck, it's a comfort, it doesn't need to be true to give it that comfort. Therefor I believe things will work out in the end. It doesn't need to be right, but I like to think so.
I love how this scientific paper merely states the absolute obvious in that scientists merely publish their perceived findings and interpretations of the data. I don't see the value or the purpose of this new paper other than to inform everybody of the obvious. 50%? It could be more or it could be less, but as far as we can tell, most are right according to our current understanding, sometimes influenced by our bias of understanding of our the world works.
Who knew!?
Uhhhh, everyone.
M.T.
"Support Bacteria - Its the only culture some people have" - Circa 1985
Meanwhile, the /. editors post provocatively-worded "articles" apparently intended to "stimulate" discussion--while the system is actively abused by moderators who use the moderation system to anonymously attack any positions they dislike. I have yet to see a legimate use of anonymity on /., though I've seen plenty of examples of abuse. If it wasn't hidden by the system, I'm sure there are plenty more examples we simply can't see.
As regards this particular article, what it really proves is that the word "scientist" is being used increasingly loosely these days. If you can't claim to be a "scientist", who's going to listen to you? Actually, if you can't make the claim, you just set up a think tank for your pet prejudice and find some "scientists" who are for sale, preferably cheap.
Finally, let's consider whether or not ost /. articles are wrong. Actually, that's a silly question. If we want to consider the question as it applies to /. articles with inaccuracies, the question is 95%, 99%, or 99.44%.
Freedom = (Meaningful - Coerced) Choice != (Speech | Beer^2), and sad sock puppets' bad mods avail them naught.
Slashdot seems to have turned into some radical leftist-atheistic-anti-religion website, even on articles having little to do with religion, all the comments are about it. WTF? Attacking ID seems to be the new way to get +5, Insightful on your comments, excuse me if I think that's a little sad.
/. has went to hell, it means K5 is probably getting better again. Time to find out.
It's like religion (or lack thereof) and politics even dominates a site intended to be about technology. If I wanted to hear your comments about how Christianity is destroying the world, I'd view Kuro5hin.org. In fact, I find it likely that if
I love the way the word "science" is equated with "truth" when they don't really have much in common.
It's also funny to see the line of argument which is, "Most scientists agree X." which really means, "you're stupid and the "cool kids" think X so you're not cool, stupid."
At least the guy mentioned in the thread post is honest. It reminds me of the studies shich looked at submitted papers to "Nature" which showed only those which the selection panel agreed with were getting published. Look up a few posts in this thread and you'll see the fallacious argument that "IDers haven't published any papers." Think about that for a moment. (Not ID vs. randomness in an infinite space, that's a totally different issue.) The poster made a fallacious statement and tried to justify it with a claim that partisan publishers published only their partisan views which proves competing view are invalid.
Yeah, OK, look at the beautiful robe the emporer is wearing. If we all think about something hard enough we can make it happen. Uh...yeah.
Where's that Star Trek episode with Harry Mudd's planet and the "everything he says is a lie." "I'm lying." segment.
Somewhere I have a really good book whose title is something like 20 questions science is incapable of answering. I'm sorry, the actual title would be the best help, I know. The book illustrates the weaknesses in the scientific method. It goes far beyond the "irrepeatable" criticism of scientific method. Wish I could remember the title now. It's all about the perils of basing "truth" upon the mistaken belief that "science" is perfect and the scientific method is perfect to the point that it slips into dogma and adherents of "science is perfect/infallible" are blinding themselves.
Oh, yeah, I almost forgot. Sorry, I'm having a real ADD day. Where are the global warming is caused by hydrocarbon use/pollution folks today? Shouldn't they be all over this thread? Maybe they're hiding because of that large hurricane which happened no matter what anyone did.
Fight! Fight!
A more accurate picture of science is a B-minus graduate student who settles for his or her masters and gives up on PhD dreams.
I suggest you read Slashdot
Comment removed based on user account deletion
from my own experience, I don't believe him. I don't even understand what does "wrong" means here
Skeptic had their take on it in the last issue. In a nutshell
"Seven Deadly Sins? I thought it was a to-do list!"
And 90% of all statistics are made up.
If science can be wrong, then why trust it?
It is the only objective process for assessing facts from fiction.
In other words, if the best a scientist can tell you today is that, he might be wrong tomorrow, why even bother listening to him?
No one is forcing you to listen. You ignore the information provided by science at your peril.
So you can use science for real things, like physics and design of military weapons and consumer goods, but the rest of it is so much speculative nonsense.
Quantum physics is speculative, but you don't seem to be throwing your computer out the window.
The consequences of guessing wrong about the origin of humanity are completely immaterial to most people's lives.
Dead wrong.
Stalin believed that Darwinian evolution was just a bouguoise concept. He believed in Lamarckian evolution and directed his agricultural ministry to ignore studies that supported Darwinain evolution. Their agricultural industry suffered and people went hungry in the process.
You can't show people evolving any more than someone else can show God making something.
I can show a progression of hominid fossils leading to homo sapien sapien. The Bible is silent about these fossils.
It's immaterial, unprovable, and so why fight over it?
It may be immaterial to you, but the theory is consistent with the evidence we possess. You may not choose to believe it, but that the only thing immaterial about this discussion.
Yeah you can roll out the eliptical argument that evolution is somehow necessary for medicine but most doctors are concerned with the human species, here and now, and now plants and people are related.
Why bother? You obviously believe that the scientific method works differently for investigations related to the origin of humanity than it does when applied to chemistry.
To wit, you can get a Chem E degree and still get into Med School.
You are correct.
Just don't whine to me when you have difficulty making sense of the data you gather without using evolutionary theory.
You will never amount to anything more than a glorified technician.
I can live with that.
"Rocky Rococo, at your cervix!"
They're all wrong.
A scientific paper is just the description of an experiment. An experiment tests a hypothesis, which is "guess" based on a theory, or model. The model is never, and I mean NEVER, completely accurate. It can't be, because it's only a mental representation of the physical phenomenon, and is not equal to the phenomenon.
So, by the very definition of empiricism and the scientific method, all scientific papers are "wrong." They're just different degrees of wrong.
Get over it.
Its possible to use logic to prove things about a mathematical model but its not possible to prove that a model is an accurate description of the physical world.
... In the end its about being able to build things that do things that we want done.
A theory about the world might be proven wrong by the very next measurement/experiment. To prove a theory it would be necessary to perform all of the experiments that test every implication of a theory at all times. New measurements can support or disprove a theory but they can never prove it. In going about our day-to-day affairs its convenient to confuse the positive feelings that derive from repeated successful use of a well supported theory with the sense that the theory has been logically proven.
In addition to the unprovability of scientific theories there are additional issues that don't make it into the Jack and Jill stories about science. At any given time there is usally more than one theory that describes/organizes the facts about equally well. When new data comes in some of these theories die, others are generalized, and new ones come onto the stage.
Furthermore, it is both a practical and a cultural issue as to which theory is the dominant or textbook theory at any given time. Any theory that organizes, describes, and predicts enough of the facts with only a few assumptions and simple rules/patterns will be a useful theory. Theories that have to deal with most situations as a special case are the least useful from a scientific and engineering perspective but they can sometimes serve political and other purposes.
Good scientific theories don't explain all the data. They don't have to in order to be useful and some of the measurements are unrepeatable. Good theories explain most of the data with some margin of error that is small enough to make the theory useful.
Often times the implications of the theory are extremely complex when applied to large systems or to systems over long periods of time. Attempting to simulate a living cell starting from string theory with no approximations will take a bigger Beowulf cluster than you can afford.
Scientific theories about the world are just working models that give scientists guidance in making better measurements to form better theories to give better guidance
Misunderstandings about the notion of proof are common and costly. Galileo wanted to prove to everyone that the planets went around the sun. As I understand it the pope at the time didn't object to Galileo's teaching the solar centric model. What he objected to was that Galileo kept going on and on about proof. The modern debate about Intelligent design has its origins in this same issue. The theory that satisfies religious needs is not the one that satisfies scientific and engineering needs. Maybe if scientists and engineers shut up about proof and started teaching the real process of science then the people who think that their religion has something to do with logical proof of something or other will stop trying to subvert the beneficial scientific process.
"/.ers: block tacoda.net (see page source) Am I missing something? I looked over the page source and didn't see anything unusual. Why is that in your sig?
There are *infinite* number of ways to reach the destination from source.
Slashdot = Sarcasm
If science can be wrong, then why trust it?
And to think you posted that with a device that is arguably high technology. Gee. It's a good thing those practical thinkers at Signetics and Intel didn't listen to those shifty eyed physicists.....
So you can use science for real things, like physics and design of military weapons and consumer goods, but the rest of it is so much speculative nonsense.
Newsflash. The same people that don't like evolution only like physics when it can be used to attack evolution. The rest of the time it gives rise to uncomfortable facts like the Earth being round and the Universe being billions of years old. They'll get around to rest of the so-called "useful" sciences once those pesky life sciences have been properly re-aligned. I'm also glad people like you don't decide what is "useful" in science. After a demonstration of electrical phenomena, the Queen of England asked Micheal Faraday what of what possible use was all this nattering about electricity. He replied, "Of what use is a newborn babe." Sheesh.
Science is all about being wrong. 99% of it is long painful slogging through mucky fields of sheer wrong and trivialities to find the occaisional nugget of right. I'm mildly amazed that Scientists Can Be Wrong is a subject of discussion. This is only a problem when people who don't have any idea how science works expect scientists to be some sort of infallible priesthood. It also doesn't help when the press seizes on new research that hasn't endured years of attacks and splashes it all over the place. It is as though Firebird 0.3 is headlined as the New Killer App. The press is the worst offender in this regard.
You can't show people evolving any more than someone else can show God making something. It's immaterial, unprovable, and so why fight over it?
You can show things that reproduce really fast evolving. It is quite easy with microorganisms and it isn't too awful bad with insects like fruit flies. It's a bit harder with some fast reproducing plants and an absolute pisser with anything that takes more than a week or two to reproduce. One can still do things like genome tracing and compare and contrast with currently living things that haven't changed in a long time. It is hard to show people evolving. It isn't all that hard to show the effects of evolution on people. Unless of course you live in the US.......
There's a famous quote by Box: "All models are wrong; some models are useful". That's what science is all about -- making models, which are useful until a better model comes along. So by definition, 100% of *all* scientific papers are wrong. But some are wrong in useful ways that inspire new generations of scientists to improve upon them.
So there is roughly a 50% chance that this John Ioannidis guy's paper is wrong over 50% of papers are not incorrect. I guess he likes to play the odds.
On a side note, I would like to report that 85% of all Slashdot posts are wrong and that 71.4% of all statistics are made up on the spot.
If you mod me down, I shall become less powerful than you could possibly imagine.
Doctors say he has a 50% chance of living, though there's only a 10% chance of that.
Dr. Ioannidis (who is Greek, like me) is right: I read daily public announcements from universities, as well as some scientific papers, and I have found that most of them are unimportant, wrong or simply motivated by financial reasons (some universities must use all their expenses allowance in order to continue receiving government money). Not only scientific papers are wrong (often the result of vanity), but students dislike studying science and technology. Recently RPI President Jackson called for a national strategy to overcome this problem. USA must invest more in science, otherwise rival nations. How would you feel to see a communist Chinese flag on Mars? You can prevent this by persuading your representatives to invest more in science and technology. The first step would be to enact more reasonable copyright and patent laws. Science, like free software benefits from openess, which is now hindered by copyright and patents. Richard M. Stallman has published an article in Nature about this, and you can read it here.
The paper seems to speak mainly about research based on statistical data. Given that he is epidemiologist this may be relevant to him but it doesn't cover all fields of science equally. So one shouldn't assume that 50% of the math departments papers are wrong, not even the ones about statistics ;).
Je me souviens.
I think we are bottoming out on the recognition issue and if we can cure our education problems, perhaps a new, perhaps more dedicated amateur effort, will restore the integrity. I distrust pharma & NSF grant grabbing profs more than the oil companies...
It *appears* to me that we now suffer much greater ethical risks, more innumeracy and less high end literacy now than 30 years ago (whither the SAT). I miss the WWII generation, they seemed to give so more effort and more humility to their work and community.
Science gives us the to tools to critically assess our experience and others' claims to know what we think we know. Study hard, read voluminously, observe acutely, think critically - and don't trust any institution to think for you.
The idea of God is that you must have faith in the existence of God. If you discover that there is an intelligent designe behind evolution then you have proof that God exists hence eliminating a need for faith! Unfortunately, the concept of intelligent design is nothing more than people trying to rationalize their faith in God. The truth is that if you argue for ID you are essentially saying that you do not have faith in the existence of God because you clearly need proof. Hence.. ID is self contridicting.
So you agree with this story.. can you show us your data and cite some examples, or do you just see an abstract problem that you think exists (regarding scientific papers being unimportant or wrong) but really dont have anything to prove it? Isn't science at a point where most progress and discoveries going to be, on an individual basis, small and seemingly insignificant, but just reflects the fact that we are living in a time where science and technological advances are so complex that the progress we make has to be systematic and more carefully approached (in scale) than before?
The very president you cite... is he not creating attention in order to get more money to tackle an 'issue'?
I don't have a problem with China on Mars. How is investing in science and technology going to prevent this? Science and technology to destroy and stop China? Hahaha.
China is only a rivial nation if we continue to act as a bully towards them.
What exactly is wrong with USA not being Number 1 in everything? This is the free market economy at work - not everyone has to be doing something with science and technology.
Personally, I'd be glad to slow down a bit and enjoy life.. I don't need every new gadget or computer, nor am I really interested in going to Mars or whatever.
...there is a good chance he is wrong
Lamarck and Darwin proposed hypotheses, some of which they were pretty sure of, which turned out to be mistaken.
The issue here is errors that the authors ought to know are errors---basing a biological theory on a poor understanding of chemistry, or a lack of statistics knowledge, for example.
10 PRINT CHR$(205.5+RND(1)); : GOTO 10
Indeed it is. Which is precisely why our senses are so easily fooled: given stimuli that do not correspond to those seen in survival tasks in the EEA (Environment of Evolutionary Adaptedness, i.e. hunting and gathering on the African plains or whatever), our brains do not necessarily respond correctly.
Now explain why a creator would have built brains that are so subject to misdirection, geometric optical illusions, etc. Why would he/she/it have done so?
belief in evolution is self-defeating because on the supposition that evolution is responsible for our reasoning ability, we have no confidence that the deliverances of reason (i.e. the theory of evolution) correspond to reality.
And if a creator built our reasoning capabilities, how do you know that he/she/it programmed it to accurately reflect reality? We'd be seeing whatever he/she/it wanted us to see, for whatever reasons. You'll get less mileage out of this argument for creation than for evolution, even.
Pointing out that evolutionary theory itself can't guarantee the accuracy of our reasoning faculties (which is true) gets you absolutely nowhere because Creation mythology is significantly worse. Consider:
Prove the above statement wrong. You certainly can't invoke anything it says in the bible, because that -- or rather your memory of it -- was created six seconds ago as well. It says exactly what the creator wanted it to say, for reasons of his/her/its own.
As soon as you invoke a creator, falsifiability is utterly gone, your conclusions can be ANYTHING, and future argumentation is pretty much futile. Thus creation mythology serves primarily as a tool for a person to project their own emotional needs and desires into their own understanding of reality.
Fortunately, there are other ways to evaluate the accuracy of our reasoning capabilities than evolutionary theory or creation mythology. Sparing a couple thousand years of philosophy, I'll stick to the pragmatic argument: Reason seems to work. It gives us effective tools for functioning, ergo we're best off assuming that our intellect and reason is what it seems to be, and make use of it.
End note: Your sig links to a story about Antony Flew "converting to religion". You'll notice he's a self-described Deist: a philosophy that is in no way contradictory to any contemporary understanding of evolution, physics, or any other branch of the sciences. He explicitly states he doesn't believe any sort of revealed religion. How does this bolster any point in favor of creationism or any other branch of post-Enlightenment fundamentalist thought? The point is lost, because Flew explicitly still rejects all that.
I stole this sig from someone cleverer than me.
Yeah. I'm as disappointed every time a pro-ID poster promotes his view on slashdot. Wanna be a Buddist? Go for it-- you aren't hurting anybody. Think purple crystals help your mellow? Blessings to you.
Believe in a religion that discriminates against gays, single moms, and wants to kill "towel heads" while promoting an absurdist "right to life"? Go to hell. Want to revert scientific education in America back a few decades with the ID meme? Shut up or read up.
ID viewpoints are not scientific (as they claim to be), and are harmful. They should be shot down. Relentlessly.
Everyone is entitled to his own opinions, but not his own facts.
Universities have failed a lot of scientists in that a) those papers are the result of stupid tenure policies and b) universities often do little to promote their researchers.
Engineers often read papers to solve problems. When they know about them! (Google Scholar might fix this)
A worse problem than them often being wrong is that:
a) there is frequently no way to determine if a given paper is accurate, has mistakes, is partially accurate, is laughable, was accurate at one point but is outdated, etc etc. At least from an outsider's perspective.
b) there is no good way to stay abreast of current interesting developments - hell, there's no way to see interesting things from 20 years ago easily! Again, this is from an interested outsider's perspective.
Once or twice a year I have the luxury of spending a week or two in an engineering library for the express purpose of finding out new and interesting things in my field. I'm SHOCKED at the amount of material that is being duplicated (often badly) in industry, material that is inaccurate or poor quality, and VERY GOOD material that never sees the light of day again.
..don't panic
To bad they didn't go with an alternative article title: "Scientific Paper Says Most Scientific Papers Are Wrong"
Can people read the article before they post it? The article clearly only applies to medical or biological studies. (It has to do with the randomness on biological systems.) Chemistry, physics and related engineering studies are much better controlled. As best I can tell there are at least as many of the later..... so >50% are wrong? How?
This message was brought to you by "Lack of Sleep."
I consider myself that articles like this should not be taken so serious (yes, I have RTFA), they do not contribute to clarify the role of science but instead throw sh*t on it. Science is built from theories, which are supported by experimental facts.n ce.html)
I recommend you read "Cargo cult science" by Richard Feynman (http://www.physics.brocku.ca/etc/cargo_cult_scie
Stalin believed that Darwinian evolution was just a bouguoise concept. He believed in Lamarckian evolution and directed his agricultural ministry to ignore studies that supported Darwinain evolution. Their agricultural industry suffered and people went hungry in the process.
Ah yes, Lysenkoism. Science and ideology do not mix well. Although, to be fair to Stalin, the people went hungry more because of forced collectivization than because of Lysenko, although the pseudo-science didn't help matters any. Ideology shouldn't trump science, social or agricultural.
78% of statistics are made up on the spot...
All women want is honesty, if you can fake that, you're in.
Scientist wrote a scientific paper saying "Most Scientific Papers Are Wrong"
.
you ID vs evolution guys need a life /.'ers an scientists, Scientific papers are ment to help the scientific community grow Thu new ideas. Even if 50% of all scientific papers are wrong, we get new ideas of what to do, how to do and can we do what we do. so what if that cure for Cancer proved false, the scientists did what scientists do, study, dream, hyophotise and test. and test. and test again.
Papers are never ment to say "OMG! I FounD A N3W 1@w OF 5cInc3!". as expressed by other, more experienced
A good study will have well documented methods and follow certain standards. Peer review is supposed to filter out poor, non-repeatable studies, although so many studies are simply observational/questionairres/etc., and don't quite fit with carefully designed scientific studies.
Personal;ly, I think if a study is wel designed and repeatble, then the data collected will generally be essentially true. A great example is the studies which showed that the disease caused by tobacco mosaic virus is in fact not caused by a bacteria. Another example is the carefulk work done by Mendel upon which he designed the theory of Mendelian genetics.
However, the conclusions drawn, even from high quality data are much more qualitative, and they are subject to the general scientific community's biases. Mendel's ideas were ignored for years, until others began to observe genetic, only to uncover that Mendel had already done the basic work.
Conclusions drawn from poor quality information can be anything from at best cautious gudance on future studies, to complete balderdash. A great example is oberservational studies that sugegsted all women should be on hormone replacement therapy after menopause. You weren't a good doctor unless you got every woman on it. Politics and women's groups certainly pushed this. However, once high quality, randomized controlled studies were in place, it became clear that HRT was in fact hazardous to womens' health. Now the US Preventative Task Force has issued a statement strongly discouraging HRT. The whole circus started with a poorly designed study with conclusiojns whch over-reached the quality of the data. Finally, a media frenzy whipped up the population into starting HRT.
Caveat Lector (reader beware), whether you're reading slashdot or the New England Journal of Medicine, you need to take things with a grain of salt and think over the data and conclusions. Remember, god only speaks to prophets and the insane. Don't assume that he was speaking to the authors of the last study you read!
Building a healthy future; Connecting communities
If you read a chem journal article, you can prove them right or wrong by repeating their experiments. They seem to be better than 50% correct. I would give them a much higher percentage of correctness.
Biologists are crazy, because they won't give up on older ideas that are wrong.
The author (whose name bears an odd resemblence to his university) is only talking about statistical studies. Something all of us knew, which is probably why we all avoided the social sciences to begin with. So you can all go back to your homes. Nothing to see here. Quantum mechanics still works.
And to answer the question of the guy who asked why we should still listen to scientists: because science WORKS. In true science, theories are testable, and nature is the ultimate BS detector. We know quantum mechanics is correct to within certain energy bounds not because of the peer review process, but because it correctly predicts the results of countless experiments. This is fundamentally different from epidemiology, which is the subject of the original article, because the theories of epidemilogy (like don't eat butter--no, wait eat butter) aren't testable.
We all knew this to begin with. This isn't really news. After the fourth time the medical profession reversed itself on whether or not it was good to drink wine, I pretty much quit listening to the cheese warning correlation monkeys.
And you all fell for it!
It is a complete joke, and so obviously bogus I can't believe anyone with a BS or MS would even take it seriously.
When viewed as satire, it is a freaking hillarious paper. I bet The Onion staff wrote it.
https://www.accountkiller.com/removal-requested
Ask them. To most kids, science is a class they take, where they have to regurgitate "facts" like why the sky is blue, or how hydrogen and oxygen combine to create water. It's a boring class, unless you happen to sit next to an attractive member of the opposite sex, but then, it's still not the class that's interesting...
Science is not 'fact' - Science is the best-known process by which truth can be reliably found.
Science is somewhat like the mathematical function x=1/y. Forever approaching truth, never (exactly) reaching it, forever leaving curious minds with new things to explore. Science is the magical combination of "what if" combined with the "feet on the ground" of experimentation, independent scrutiny, and validation of theories.
The "Scientific method" that is regurgitated by most Jr. High schoolers (in California, anyway) is never really *experienced* except in the case of the rare instructor who goes above and beyond the textbook curriculum. EG:What drudgery! If that really was science, I wouldn't be interested, either!
It's sad. Entire generations of people who never get to experience the awe, wonder, and magic of science, who then seek out that awe and wonder by (best case scenario) watching magicians and listening to Art Bell, or (worst case scenario) performing criminal acts and doing drugs.
How much of the interest in the pseudo-sciences (aliens, conspiracy theories, perpetual motion machines, telepathy, Scientology) comes from the fact that they have never really been exposed to the real thing?
I have no problem with your religion until you decide it's reason to deprive others of the truth.
The problem is not that many scientific theories are wrong, we all knew that. The problem is that a majority of published scientific papers are provably wrong at the time of publication, and the author should've known that it was wrong, but is too stupid and/or busy to publish a correct one.
Scientific papers are usually written by grad-students trying to earn a degree, and that is usually the only real purpose they will ever serve. The project I am working on now is a continuation of the work that was carried out by someone who now has a PhD. Nearest I can tell, one of the important equations he used was not appropriate for our equipment. It's just a +/- error, but it's a pretty big deal in terms of the data you get. He also made some rather inappropriate assumptions. A paper was published from his work.
People need to realize that "scientific journals" are simply catalogues of the work that has been done by various grad-students and do not necessary reflect reality. I'm not saying they're not useful, I'm just saying that they aren't often correct.
Speaking as one who has read his share of "questionable" papers in the life sciences (and also knowing the questionable analysis the authors of this "paper" used) I have to say I would not be SURPRISED if a significant number of papers were "wrong". That being said, one has to understand what I mean by "wrong". Rarely is it the case that papers are blatantly incorrect; usually the case is that the data were misinterpreted, or that the authors attempted to make a broader generalization of their results that is not supported by their data (the latter is the most common situation). In journals (especially big ones: Science, Nature, Cell, NEJM, JAMA, etc), authors are usually under pressure to show an interesting "story" of their research (these are big deal publications, after all). Reviewers press for things with "broader biological significance" than "the factor X is involved in modulation of this pathway under condition Y"--usually they go more for things like "the factor X is a critical regulator involved all biology, ever". So authors read further into their results, and shoot out grand theories and models and "universal paradigms". If nothing else, it makes for compelling literature. But the truth is, most of the time, the results and immediate conclusions made ARE well-reasoned and reliable. It's just the "broader significance" that tends to be overestimated or poorly backed up.
This is just like a regular Open Source project's documentation! 50% of it is false or just too old to be used for anything. It should be read only to get ideas on how the hell the project's applications should or can be used.
This guy is listed as being an "epidemiologist" which is in medicine (the mentioned senior editor is a neuroscientist - MD). IMO, these people aren't (nor have they ever been) involved in actual science. They poke and prod and see what happens and report there findings.
/sounds/ right and goes with it. Ooooooo, I'm confident.
How is there supposed to be peer review with this type of research? Is the referee supposed to go over all there data and analysis and... No, they just have the paper, and with out the above, a proper referee report cannot be generated. So, the editors just pick what
Now in Physics or Math or one of the "hard" sciences, they actually do something that can be checked (at the paper level)! They describe something repeatable regardless of who does it. Or in the case of math (theoretical physics, etc), verify the proof.
Try doing that with people ie The reaction to drugs varies from person to person and even for one person the different times they take the same drug. Good luck!
This "conclusion" might be true for MD's (I can't verify/falsify), but this can hardly be said for science in general. How can people generalize from one field to all? The answer is that they can't.
I'd say that this guy engineered these "statistics" to say in the end what he wanted to say from the beginning. It's actually quite easy to do with stats and can be quite hard to tell when done. In fact, it happens all the time whether the author is doing it intentionally or subconsciously. It's what makes stats so dangerous; you have to be quite careful, which I didn't see in the paper at all.
He actually just accepts what the other papers have lead to without question. Kind of ridiculous given his claim that 50%+ of all papers are wrong.
I mean, if what he is saying is true, then over half of what he is referencing to prove his point is false. Thus this invalidates his claims which basically turns this paper into the lying paradox (This sentence is a lie).
We don't even need stats as we already have *years* of papers and results that can be used to figure out an actual number. Why not take all the papers from the 70's and 80's, count them and count how many turned up true/false. Calculate the ratio and bingo bango, you got a true number.
And since this really isn't all that long ago, and it is over a good span of time (it could be made longer as well), it should be a good approximation of what is going on currently. It'd be nice to have this true number instead of a paper that (which by its own claims) is probably false.
But then again, that'd have to be done field by field. A lot of energy spent. I doubt if the required amount of people/time/etc would be willing to expend that much energy.
My conclusion is that this article/paper, and its misguided "conclusion," are terribly misleading to the point of being flatly wrong.
If he'd limit his conclusion to his own field, then and only then, would I consider it.
Maybe he put the "o" in there and it removed the accent, just like it did for me too. Damn you, Slashdot!
Karma: It's all a bunch of tree-huggin' hippy crap!
... a randomly chosen scientific paper has less than a 50% chance of being true. ...
So I'm guessing he released the paper at least twice.
Wolfgang Pauli's comment on one scientific paper shows that there are worse things in science than just being incorrect. Science is always falsifiable.
The funny thing is that this is only a debate in USA.
Someone has found a way to blame the Americans for hurricane season. Bravo gentlemen, Bravo.
Oh, and you're sure being a beacon of insight, logic, and rationality.
Information wants to be anthropomorphized!
Ok, as from someone who is in thie "business" of research, and papers "creation", you have to know, that there is no perfect idea, there is no perfect solution, there is no perfect paper. But this is not the goal, either. Conferences and conference papers are there to provide a ground for scientists to make their latest stuff public and let it be chewed and digested by others. It's after many iterations and discussions and quarrels sometimes, when one either gets to a point when the re- and re-corrected idea seems to work ok, or it turns out to be useless junk although it seemed like being good at first.
:P
I read many papers, I don't know numbers, but many dozens in a month. Usually I don't care if they seem good or bad, if they are correct or not. The ideas therein are what matter. Sometimes you get ideas on how to improve an old idea, sometimes you get new ideas from older papers. sometimes it's just nice to know what others are doing.
The matter is very much different when you have to review papers, but the seriousness of that review also depends very much on how much time you have, possible IRL problems, etc., but that's why there are >=3 reviewers+associate editor assigned to the paper at most of the serious journals.
Stating such things as a certain percent of all papers are crap is just crazy sh*t. It happens very seldom that I read a published (conference or journal) paper and I think it was useless. Anyway, if it would be true that would mean that this guy's paper is also half useless. You are free to choose which half
I am putting myself to the fullest possible use, which is all I can think that any conscious entity can ever hope to do.
it's like today's sig
A pretty woman can do anything; an ugly woman must do everything.
a science paper is like a pretty women to some people and an ugly one to the rest!
http://thearbitcouncil.blogspot.com/
It's not that scientists shouldn't have to publish their research, but that we should value research that did not yeald the results we hoped for higher.
For example, a study that basically says "I thought this substance could be useful against cancer. So I incected it into cancer cells to see if they died. They didn't" is actually very useful research.
Scientific theories are like software components: they keep getting upgraded. Sometimes the upgrade is a patch that covers a few more corner cases. Sometimes it's a complete rewrite to a new framework. As with software, the upgrade usually comes after additional testing.
My point is this: in both software and science, the old theory/component has value. It's not wrong in the sense of useless, and it doesn't lose value after the upgrade. The value of the old theory/component is unchanged; if it works in your use case you go on using it in your conceptual framework/program.
In other news, the ozone layer is not shrinking today.
Um, This article is kinda boring and contains very little content. Why the hell did it make the main page of slashdot.org? Is nothing cooler going on right now? I regret that I cannot report more interesting stories myself, but I sleep in a small bubble without steady supply of electricity or water. Those aren't slashdot stories. ciao
Stalin believed that Darwinian evolution was just a bouguoise concept. He believed in Lamarckian evolution and directed his agricultural ministry to ignore studies that supported Darwinain evolution. Their agricultural industry suffered and people went hungry in the process.
That statement makes no sense whatsoever. So a nation has to believe in evolution to feed its people? For lack of better terminology at the time, that's stupid.
Stalin was a communist murderer. Communism, on a national level, is never voluntary. Stalin purposely withheld food to kill off people who didn't want to "join the group." The amount of people Hitler killed was a drop in the bucket compared to the number of people Stalin killed.
Usurper_ii
Ron Paul
"Slashdot editors post exponentially increasing number of dupes"
Well, this guy is an epidemiologist, not an epistemologist. So I cannot understand how he write a paper on a problem on which he is not competent. Maybe epidemiologist publications are all false, but not scientific publications. When I see how long it takes for a scientist to write its article, to check it for errors, I cannot understand that such a man says that they are all wrong ! Ok, some can be, that's science. But not 50% of scientific publications. And what does he call scientific publications ? Publications in "Nature" or in the "Scientific American" or in "Science for the n00bs" ?
Bonjour !
I think the important thing to take away from this is that people need to keep science in perspective. It seems, though, that a high percentage of the Slashdot crowd have created the Church of Science (insert Tom Cruise joke here). There is only one true Science, and Darwin is its prophet. Right.
I can't be the only one who reads this message board and is chagrined to see the religious fervor that's incited whenever anyone challenges evolution. I also can't be the only one who realized that, while the TFA said NOTHING about evolution, the very first post on this page was basically an apologist saying that, while the data might be wrong, that evolution is still right, when NO ONE had challenged it. Maybe some psychologists among us can explain that.
Couple that with the following quote from the article:
small sample sizes, poor study design, researcher bias, and selective reporting and other problems combine to make most research findings false
The key one here is 'researcher bias', which leads to 'selective reporting'. The scientific method guides scientists to start with no preconceived notions, gather as much data as possible, and try to interpret what the data points to. However, I fear that evolution, if it's wrong, will take a long time to fall to new ideas, since data that doesn't support it may be overlooked if it doesn't fit the mould, or suppressed if the researcher is afraid of reproach or ridicule from his peers. I would suppose it's very difficult for scientists nowadays to do objective research on the topic because of this.
I also submit to you that, evolution, true or false, does not prove or disprove God's existence. It's a non sequitur to believe otherwise. To prove an assembly line exists doesn't disprove the existence of a widget's designer, it speaks to his ingenuity.
The vogue on Slashdot is to declare that God is Dead (or Never Was) without providing any evidence for such a daring statement. The general complaint is that God is not falsifiable. That just makes the science of finding him difficult, but, again, doesn't speak to His (non-)existence. To say that it's impossible for Science to find God speaks to a deficiency in Science - not in God. But Slashdot unwittingly reinforces the notion of the article, namely, that Bias takes the place of Reasonable Conclusions in too many cases -- especially in cases where the Reasonable Conclusion is that a reasonable conclusion can't be drawn.
"That statement makes no sense whatsoever. So a nation has to believe in evolution to feed its people? For lack of better terminology at the time, that's stupid."
Except the grandparent didn't say that. He said that that Stalin couldn't feed his people because he directed his ministry to ignore studies that supported Darwinain evolution, which meant his country used inferior methods. The grandparent did not in any way conclude that a nation has to believe in evolution to feed its people. He merely pointed out that researching the origins of humans has valuable and real benefits beyond knowing the truth, and that to ignore such information can be detrimental to society.
This guy are sick.
God is perfect.
God is infinite and has no desires.
The Universe exists because God exists.
Both God and the Universe are infinite.
The Universe and God are the same thing.
....78% of all statistics are completely made up garbage.
What's the entry level now for a post to make it to slashdot?
"Champagne for my real friends - and real pain for my sham friends!" http://ericblade.postalboard.com/
Quantum physics is speculative, but you don't seem to be throwing your computer out the window.
The predictions of quantum field theory are empirically supported. It is anything but speculative.
Even in the journals that I regularly read (every issue, every year), I only read a relative handful of papers, germaine to my research. When my research topics evolve, I might go back and read different papers in the same issue. Maybe there are some scientists out there who read every paper in every issue of journal in their field, but they must read a hell of a lot faster than I do. I rely on Current Contents, automated lit searches, and other computer-based tools to sift through the flood of info. I also rely on my colleagues - they know what research I do, and I know what research they do. If I see something that might interest them, I forward it to them, and vice versa.
there is no good way to stay abreast of current interesting developments
I would respectfully point out that that's why the annual scientific conferences are useful. The research presented in the talks and the posters precedes that presented in the papers and book chapters, giving you a feel for what the latest interesting problems are. If all of a sudden there are three times as many posters on Probelm X at the 2005 conference than there were at the 2004, then that should tip you off that something is up. If they are all coming out of one institution, that should tell you something, too. I know IEEE and other engineering societies hold annual meetings; are they not as useful as, say, ASM?
Once or twice a year I have the luxury of spending a week or two in an engineering library for the express purpose of finding out new and interesting things in my field.
This means that you are necessarily reading the journals at least a month, perhaps as much as a year after they come out. I don't mean to flame, but I suggest that this is not a very good strategy for staying current. By the time something is published in a journal, a lot of people will have known about it for a year or more, right down to the experimental details.
As to how to tell good work from bad work, that's what collective and individual professional judgement is for. If the profession is divided, and your individual level of expertise in that particular area is inadequate to make a good judgement, that's when you ask a few of your colleagues, "So, did you see that presentation on Problem X from Dr. Smith at Big State University? It looks like he was directly contradicting Dr. Jones from Small Private University. What do you think?" If none of you can tell who's probably right, then you either wait for more data or go generate the data to decide the issue.
The man who does not read good books has no advantage over the man who cannot read them. - Mark Twain
Does that mean I should get off my low-carb diet and get back on my low-fat diet???
Evolution is a fact. Darwinism is a joke.
everyone believes global warming is a reality
Did anyone see the recent NOVA program about Rosalin Franklin. She was doing DNA research at the same time as Watson and Crick, but in a different lab in England.
It turns out that Watson used her data without her permission and without attribution. And he went on to seriously misrepresent her in the book that he later wrote on the discovery of DNA. In fact, Harvard, the original publisher, ended up not publishing the book because of the complaints about the way she was portrayed by many of the other people who were there and mentioned in the book.
Watson basically created a fictional account about the way that DNA was discovered. And the public at large drank it up.
Watson got the Nobel prize. Rosalin Franklin is hardly remembered.
The scientific community is as full of intrigue and back-stabing as any other human community. Well, maybe except for Slashdot.
Evolution is a fact. Darwinism is a joke.
- description (of phenomena)
- understanding (hypotheses about relationships)
- prediction (this requires general applicability of hypotheses)
- control (use knowledge to influence phenomena)
And the key values of science are:- Empiriscism (objective evidence!)
- Skepticism (available knowledge may always contain errors)
- Tentativeness (scientific knowledge is never the truth, but always a theory that tries to come as close as currently possible)
- Publicness (results and methodology must be published so others can replicate results, thereby establishing validity)
See how some of the basic points we're taught are also what the article tries to point out? This should be common knowledge, at least to scientists. And non-scientists that argue against science usually go against this uncertainty, as if tentativeness were something bad. "It's just a theory" should not be used in a derogatory sense. As the first post managed to (once again) get this discussion rolling down the Evolution(ism)/Intelligent Design(ism) road, here are some useful indicators of Pseudo-Science, also from my class:- Subjective
- Illogical
- Unsystematic
- Not empirically testable
- No predictive power
- Knowledge system is fixed or closed
- Anachronistic thinking
- Seeking mystery or lack of patterns
- Appeal to myths
- Casual approach to evidence
- Irrefutable hypotheses
- Explanation by scenario
- "Literary" interpretation
- Refusal to revise
To make this discussion a little more clear: "Irrefutable hypotheses" in my understanding is similar (if not equal) to a hypothesis not being "falsifiable", a term others have repeatedly used here. So basically: a theory is worthless if its hypotheses can not possibly be proven wrong by later research. This is because from such a hypothesis you can come to any conclusion. As as example to this I especially liked the 6-second universe theory stated by someone earlier in this discussion. As you can't prove him wrong, the theory is worthless to the basic goals of science.Incidentally, Richard Dawkins illustrates an important fact in relation to evolution like this:
"There are vastly more ways to be dead than alive."
617B3B7F7E7C7D7F00EOF
Ioannidis' previous publications summarize the results of clinical studies, that is, studies on human populations. These studies are generally not replicated before publication because they are long and expensive. The results of these studies are often published, in spite of the lack of replication, because they appear to be statistically significance. But statisical significance does not assure that a result is repeatable, and, in fact, many such published results are not.
However, in many (most?) scientific disciplines, experiments are replicated, sometimes many times, before publication. As you might expect, experimental results that are replicated in one laboratory before publication are more likely to be confirmed in other laboratories after publication. Ioannidis' claims do not apply to such experiments.
All this talk about truth...
Statistical findings are never true, even when they are essentially accurate. They are guesses about properties of populations that are themselves hypothetical. The problem is that the scientific community likes to portray such fundamental methodological weakness as if it were precise, replicable, theoretically informed truth. Statistical findings are, at best, weak empirical indicators that only strengthen when underlying mechanisms are understood, at whch point the statistics themselves become superfluous.
Peer review cannot overcome the problems with theoretically weak, poorly executed and interpreted statistical research--even when population parameter estimates are accurate. The peers are themselves too committed to the paper production system. If the empirical goal of identifying important problems and then tracking down underlying mecanisms is secondary to paper production--which too often is more about making journal editors feel comfortable than about the serious debate needed for scientiic advance--then statistical science becomes mostly a fashion show. I think the author of the article is onto this and, yes, many weak scientists probably should feel threatened and probably would not be happy with this analysis. They have no sensible argument against it. Paper production is only the goal when you are not advancing as a science. I would venture to say that the more papers in a field by more people blindly espousing the statistical empirical ideology --while never actually following a problem to a more definitive conclusion-- the more likely we have a science that is not advancing, and probably cannot advance. Reporting scientific advance should be the goal, not numbers of papers produced. But papers produced is the coin of the realm in what is presented as legitimate science. Thus, much of this corpus of "scientific literature" is trash.
I am very glad the author of the paper being discussed has raised these important questions.
What you describe is what happens in very small populations. In more moderately-sized populations changes are driven by a more even mixtue of drift and selection. In very large populations genetic drift is all but negligible, and selection predominates.
10 PRINT CHR$(205.5+RND(1)); : GOTO 10
At least as I've heard it used "genetic drift" does contribute to evolution; it's essentially the "sampling error" factor in the stochastic process. That is, if you have 5 individuals each of A and B, and the ideal result due to selection is that the next generation ought to have 5.5 individuals of A and 4.5 of B, getting 4 of A and 6 of B just due to random chance is not that unlikely.
The OP is wrong in that this is generally the phenomenon that dominates though; that's only true in small population groups, where the variation from sampling dominates most selection pressures, which tend to be relatively modest. In large population groups even small selection pressures tend to dominate.
10 PRINT CHR$(205.5+RND(1)); : GOTO 10
The problem I see with this isn't the scientists, it's the general public. How often have we heard (shotty) news stories on "ground-breaking studies"? And of course the general public just eats it up! The average person doesn't spend the time to check every study, and see how valid it is. This could be dangerous (I just can't think of any examples)
I'm as disappointed every time a pro-ID poster promotes his view on slashdot.
/.
Name once that an ID proponent made a comment on a NON-ID related story here on Slashdot. You can't. Yet the examples of atheists posting jokes about how "dumb" ID is on articles not even related to the debate are so easy to find because they are +5 modded all over the place.
I don't even like ID and I find it ridiculous that far left radical atheists have hijacked
After all, where do you think all those intra-species variations came from except mutation, interbreeding and natural selection?
I think the point was that natural selection does not produce the variations. It is the determining factor of which variations are successful. Thus you need both variations (which are passable to offspring), and selection (of those offspring) to describe the manifestation of organisms present today.
We must remember that evolution is defined to answer the question, "How did we get all these different species of plants and animals?"
And when you get right down to it, selection isn't needed to differentiate species (unless you consider incompatibility between species to be "selection" rather than the determining factor). It is needed to weed out species.
So as mutation sews the seeds of variation, selection weeds out unsuccessful varieties, and promotes "desirable" characteristics.
Without natural selection random mutational variation would eventually populate the universe with a near-infinite number of completely different individuals.
While we don't exactly have a "near-infinite" number of completely different species, we do have a great diversity. If selection was more important than mutation, then we would have only one (or much fewer) species.
Mutation is why we have so much variation, selection is why we don't have a continuum.
natural selection is still the prime mover (in fact the only mover) in evolution.
You could have evolution without selection, but not without mutation. A variation that does not exist cannot be selected.
Certainly, both are important to evolution, but it would be more correct to say that variation drives evolution, while selection shapes it.
Basic statistics is not that hard, but getting things right does require a high level of clear thinking. Unfortunately, too many people don't understand the basic principles. Even worse, there are computer programs which enable people to use all sorts of statistical techniques, including the more advanced ones, without really understanding what is going on. Unfortunately, it seems to me, many scientists prefer publishing lots of 'soft' papers to taking the time to do rigorous statistics. This is only to be expected, because the incentive structure for academic careers tends to reward quantity of publications over quality, or is at least perceived to do so. (This is part of a much bigger and harder problem. Systems such as peer-reviewed journals worked well for the much smaller pre-WW2 academic communities, but seem inadequate for today's needs.)
Of course, it's not just scientists who have problems with misunderstanding and misuse of statistics. Perhaps we should regard "statistical literacy" as a fundamental attribute of a good education, along with "computer literacy".
Speaking of which, can anyone recommend a good book on statistics for non-dummies? I think I should do some reading ...
It's the way you typically model bacteria populations, for example. Or even, on a slightly smaller scale, insects.
10 PRINT CHR$(205.5+RND(1)); : GOTO 10
Everything in the Universe is caused by something else, requiring a cause to exist.
One imperfect thing may be caused by another, but the causer needed to have been caused by something else (since it is imperfect and requires a cause).
If imperfect things exist, there must be a being who can cause them, but having the characteristic of not needing to be caused himself (and this being we call God).
If there were no God, nothing imperfect, requiring a cause, could exist.
But imperfect things exist, therefore God exists.
"That's a load of malarkey! The earth isn't just floating out there somewhere; it sits on the back of a giant elephant."
The scientist looks at the man and asks "So what is the elephant standing on?"
"A huge tortoise."
"And what is the tortoise standing on?"
"An even BIGGER tortoise."
"And what is THAT tortoise standing on?"
The old man thinks for a moment, then smiles wryly at the scientist. "You're very clever young man, very clever. But it's tortoises all the way down!"
Any sufficiently simple magic can be passed off as mere advanced technology.
No one is forcing you to listen
But au contraire, we have a public body that demands political action be taken in defense of one or more results that may be flat out wrong.
If you want to make actionable law out of what you do, be right, otherwise, don't bother us.
Quantum physics is speculative, but you don't seem to be throwing your computer out the window.
Don't go linking evolution to quantum physics. Quantum physics built me a computer. I've yet to see an evolutionary biologist create a planet out of dust and produce a human being.
Dead wrong.
Completely true. Name me one consumer good produced by evolution? I can name thousands produced by physics.
I can show a progression of hominid fossils leading to homo sapien sapien
And so what? Theoretically I could write a computer program that ties together every known fact of evolutionary theory and creates an infinite number of internally consistent systems around those facts. You cannot predict with your rocks, cannot repeat any steps, cannot test your ideas by repetition and cannot prove anything.
You will never amount to anything more than a glorified technician
And you'll never be a scientist, only a slave to inference. TEST. TEST. TEST. REPRODUCIBLE RESULTS. That is science.
I'm not by any stretch an ID guy or a creationist whack job, but I'm sick of people polluting science with politically motivated and untestable ideas. If you cannot test something, it is not a fact. Period, end of story.
This is my sig.
And to think you posted that with a device that is arguably high technology
Yes, that is based on test, actual experiment, and reproducable results. Computers are real science, real fact.
Newsflash. The same people that don't like evolution only like physics when it can be used to attack evolution
I can't help it if they are idiots. My whole point is that the sales pitch science as a product is flawed because it presently treats inference as fact and thus, because it says its "facts are wrong", causes people to incorrectly invalidate the whole lot.
All of science can be distilled as a collection of data points. Assume that for a given domain, we have two points. Mathematically speaking, we know that there are an infinite number of functions that can satisfy those two points. A model in science attempts to predict what is in between those points and quite often that model is flawed or wrong because it guesses incorrectly.
So, when we say we know something, we need to be clear and communicate that we only those two points as fact, and that we INFER that a function of those two points which we can test. But at the end of the day, the only thing we know as fact will be those two points, and each new point we test will be just one more fact.
expect scientists to be some sort of infallible priesthood
Scientists sell themselves as that. And they aren't, and people are learning that, and they are wrongly dismissing all of it as crap.
You can show things that reproduce really fast evolving
Of course you can. And you can make new products by tinkering with genomes. That's a data point for you. But it is still an inference to say that people evolved from apes (which I think we did), not a fact.
Facts are things you can reproduce. Anything else is inference. Because science mixes the two in its public message, people throw out both, incorrectly.
Really, if you view the human enterprise of science as a sort of a brute force search algorithm of the unknown, one might think that we would teach people all sorts of random but curve fitting models so that they will think about things differently so problems can be approached in parallel from many different angles at once.
Paradoxically, to even teach everyone the same models might actually be slowing down scientific progress for new and better models.
This is my sig.
>evolution and directed his agricultural ministry to >ignore studies that supported Darwinain evolution.
That's actually not true. Stalin cut off the food to parts of his country friendly to the white russians to kill off political opposition. Everything else he made up to cover his tracks, even the crap about Lamarckian whatever.
This is my sig.
Who the heck moded the parent a troll?
TEST. TEST. TEST. REPRODUCIBLE RESULTS.
Got a test for every theory?
How about gravity?
I guess you are back to technician.
Thanks for playing.
"Rocky Rococo, at your cervix!"
TEST. TEST. TEST. REPRODUCIBLE RESULTS. That is science.
Well, yes; that's part of science. But consider astronomy. It's widely considered one of the hardest of the "hard sciences". But astronomers do rather little testing or experimenting with their their objects of study. There are certain technical details that prevent this. Reproducible results are right out, for the most part, if you can't even do a single test during your lifetime (or the lifetime of your species).
Of course, astronomy is what we call an "observational science". It's not as fast as some others where experimentation is possible. You have to have lots of theories and hypotheses waiting in the wings. Then when an interesting observation comes along, you put it up against all those theories and hypotheses, hoping that it'll disprove some of them. Then you sit watching and waiting for the next interesting event.
Lots of astronomers would love to be able to experiment with their subject matter. Even more, they'd love to be able to repeat the experiments and get reproducible results. But they probably still have a bit of a wait until they can accomplish this.
This doesn't make them non-scientists, or slaves to inference. It just means that they spend a lot of their time in data-collection mode. Or, more and more, they figure out how to make machines do the data collection and basic analysis.
Astronomers aren't the only critters to use this approach, of course. The web-building spiders have a similar hunting strategy. Build a trap for the prey you're after, and sit by the trap waiting for it to be triggered. Dash over, grab what was captured, and devour it.
It's an adaptive strategy for some.
Those who do study history are doomed to stand helplessly by while everyone else repeats it.
I'm mildly amazed that Scientists Can Be Wrong is a subject of discussion. This is only a problem when people who don't have any idea how science works expect scientists to be some sort of infallible priesthood.
Well, I think it's because, despite how often scientists are wrong, during the past couple centuries of the scientific explosion, they've been right so much more than anyone else that it has given them this sort of undeserved reputation.
Thus, humankind suffered for millenia with scourges like smallpox. All the best efforts of the religious people had no effect at all. No matter how much they prayed, we kept catching the disease and ending up either dead, disabled or disfigured by its ravages. Then William Jenner came along, the disease was pushed back in first a small part of the world, then in more and more areas, until now it exists only as a few speciments frozen in liquid nitrogen. Unless someone does something really stupid, nobody will ever again have a smallpox-vaccination scar like the one on my upper arm.
Meanwhile, the same story has happened with lots of other diseases (though not all of them). We're looking at the eradication of measles and polio, which would have happened already if it weren't for the intercession of some of those religious people in several countries.
But the really annoying part is that scientists and medical people have long been warning of the dangers of indiscriminate uses of antibiotics. Disease organisms will evolve resistance to the drugs, they said, and we'll lose the ability to control the diseases. But the religious folks suppressed the teaching of evolution, and people like farmers and feedlot owners used antibiotics because they didn't know why it was a bad idea. Here in the US, there are large sales of "antibacterial" soaps, which people buy and use because they went to schools that weren't permitted to teach evolution, so they don't understand why this is a really bad idea.
And guess what? Lots of disease organisms are evolving resistance to those antibiotics. Now we have a number of diseases (e.g., malaria) reappearing that we thought were under control, and our controls no longer work.
But still, the religious people insist that God (or the Intelligent Designer) made those organisms susceptible to those antibiotics, and evolution doesn't happen, at least not on time scales that we can watch, so that can't possibly be why those diseases are spreading.
Intelligent people wouldn't let them get away with this. Evolution is happening right before our eyes. You can see it in hospitals around the world, where people are dying of diseases that we had under control not long ago. They're dying because disease organizations (and vectors such as mosquitos) have evolved resistance in only a few years. In some cases, there's a new antibiotic that will work, but it's patented, so the maker can keep the price too high for most of the world's people, and they die with the cure out of reach. But if the price were low, we'd just overuse it, the organisms would evolve resistance, and people would die anyway.
The really annoying thing is seeing people continually argue that evolution is some sort of esoteric, academic, intellectual topic. It's not. People dying from a disease that was once under control is not at all an intellectual issue.
And note that scientists do understand this process. The problem is getting people to listen, when their religious leaders are telling them a different story, and suppressing the teaching of biological reality.
Those who do study history are doomed to stand helplessly by while everyone else repeats it.
Your post and the parent post to yours are both absolutely excellent and spot-on. Thank you, it is refreshing to read intelligent posts from other people in the scientific field.
It is precisely the fact that science theories can be wrong, that makes it science. If there is no possibility that the theory is wrong -- that you must accept it on faith -- then it isn't science.
Science is all about "cause" and "effect". If all you have is "effect", and the "cause" must be accepted on faith as undiscoverable, then again it's not science.
Science is about finding the simplest explanation that fits all the known observations. When we find an observation that doesn't fit the current explanation we must conclude the earlier explanation was "wrong" and needs to be modified, or abandoned.
Science rejects the notion of a mysterious unknowable cause. Science abandons it's errors. That's what makes it trustworthy.
Science is not about good or bad, ethics, religion nor morality as these are not natural phenomenon subject to experimental testing, there is no cause and effect.
- AndrewN
If the standard model says that gravity is caused by the exchange of gravitons, then we can build a big enough particle accelerator to go look for them.
I guess you are too big of a wimp to be a scientist.
This is my sig.
If the standard model says that gravity is caused by the exchange of gravitons...
Oh, I see. So where is the demand for:
"TEST. TEST. TEST. REPRODUCIBLE RESULTS."
You tone is less demanding than when you are not discussing the origin of humans.
I guess you are too big of a wimp to be a scientist.
Coming from someone who exhorts disdain for the scientific method, I take that comment as a badge of honor.
You are Jerry Falwell make good company.
"Rocky Rococo, at your cervix!"
If you actually read the article about "most studies are wrong", you see that it was about finding statistical significance in studies about populations (mostly in medical studies.)
The studies about the ozone layer do not fall into this category.
The "most studies are wrong" article was an interesting view of a statistical phenomena - if you are looking for positive correlations with statitics at a 95% confidence level, your incidence of false positives may be higher than 50%, for a sufficiently low number of actual positive correlations.
That's very relevant to medical studies, but not to observations of the ozone hole.
References to that article are going to annoy me for a long time, I can tell.
It's not wasting time, I'm educating myself.
The best argument against ID I ever heard was basically: Science is the study of the natural universe. ID is the proof that their is a super natural. ID is therefore not the study of the natural universe. So, ID is not science. ID is just logic.
ID isn't "proof" of anything, ID is conjecture, pure and simple. ID isn't testable or falsifiable, therefore it isn't science.
Granted, many people go way overboard in refuting it. These people's rhetoric often degenerates quickly into personal insults. But they are basically correct: ID isn't science because it's not testable or falsifiable.
And I'm curious about your statement that science is a "limited field". The only limit it has is it will only accept that which can be either verified or falsified. It does not accept conjecture or anything that is based on faith rather than verifiable evidence.
Is that what you meant?
/.: why the hell am I here?
The scientific method means test and reproducable results. Anything less is not the scientific method. You are the one that wants to open the floodgates of inference and allow nonfact to pollute the knowledge base.
Now you are just mixing arguments. You can test the standard model's theory of gravity. Until that time, the cause of gravity is anybodies guess. At some point, we -will- test the standard model's theory of gravity. Until that time, it is not a fact. Just an inference.
Sorry that you can't see the difference between making stuff up and not making stuff up.
This is my sig.
The scientific method means test and reproducable results.
Are you back to rehash your lack of understanding?
The scientific method is based on observation and experimental tests. Theories (as opposed to wild-ass-guesses) are explanations formed on the basis of experimental and observational evidence.
My eleven-year-old daughter has a better grasp of the scientific method than you do, Todd.
Anything less is not the scientific method.
And you have proven, yet again, that you haven't any clue, Todd.
You might want to delve a little deeper into the foundations of science before you dig yourself deeper into a hole of misunderstanding. I would start with the philosophers of the 17th Century. Work your way back to the present before you make an ass out of yourself - again - Todd.
You are the one that wants to open the floodgates of inference and allow nonfact to pollute the knowledge base.
Yeah, like God is verifiable. Where is the 'fact' in that statement, Todd?
I guess you also have no idea of the meaning of "faith".
Sorry, but you are just retreading the same old tired arguments - some dating back 400 years.
Those arguments were found to have no scientifc foundation then and continue to lack scientific foundation now.
The funniest thing about this colloquy is that you actually believe that you have an original argument.
Sorry that you can't see the difference between making stuff up and not making stuff up.
So fossils are made up?
Sorry, but you have lost again,...
Todd.
"Rocky Rococo, at your cervix!"
Did you really mean this:
"It will be a cold day in hell before I feel sorry for the victims of any dictatorship ever again."
Wow, that is pretty cold. I'm sure that there are millions of Ashkenazi who would violently disagree with you.
Hell, your attitude was clearly reflected by Americans before WWII when a boatload of Jewish refugees arrived in the US. The authorities turned them back to Europe, and certain death. Good thing we have learned from that mistake and now try to recognize these tragedies before they exterminate millions of people. We haven't been entirely successful, but we still have folks like yourself who frustrate our efforts to save lives.
I now know why you have such a piss-poor understanding of science: You have a limited historical perspective. Try sitting down with victims of the Holocaust and make the same argument about tolerating dictators. You might also parade your lack of sympathy for the victims again and see if they let you get to the door.
History, science, foreign policy, global epidemics,...
There is just so much you are wrong about.
"Rocky Rococo, at your cervix!"
Ok batman, here's the point you miss. Science is a tool to discover and share knowledge about the real world. Therefor, it has two components, it has knowledge, and it has the real world.
For the real world to be satisfied, science must interact with it.
For the knowledge to be satisfied, it must be shareable with others.
For this reason, we combine the two requirements into a single principal - we test something about the real world, in such a way as that someone else can repeat the same test and get the same result. This way, the knowledge can be communicated, a real world result is achieved, illustrating that knowledge, and multiple people can get the same result given the same test, illustrating not only the truth of the matter but also that the communications itself is valid.
Thus, experiment against the real world is what science is about.
The whole foundation of this method occurred because once upon a time many people felt that all knowledge could be inferred, by process of logical reasoning, against the known works of the ancients. There was no need to test things themselves, because, if the logic was sound, therefor, the fact must be true. Jonathan Swift, in his famous Battle of the Books, argued this point, ultimately on the losing end. Galileo wrestled with this in his little "play" that got him into so much trouble with the Church. Newton in some ways arguably put it all together with Optiks - not only did he come up with the math to explain things - he also had the novel idea of checking his math with actual experiments. Soon, other people did that as well.
Experiment is everything. We don't teach kids to measure the temperature of boiling water just to verify that water boils at 100C. No, we do that and many other experiments because we inculcate our children with the idea that if there is a fact presented to you, then, you should be able to check it or test it yourself, otherwise, you can freely reject it.
So.. if you can't check it, or at least, trust in someone who can check it, argubably, it isn't a fact. So, we can say that yes, evolution of bacteria or fruitflies is a fact, and from that we infer that man evolved - but, that does not make the evolution of man a fact, only a good inference, because we have no test that we can repeatedly evolve a man.
There is absolutely nothing religious about this argument. As I said before, I reject so called Intelligent Design and creationism because it too cannot be tested. All I am saying is, science should not mix fact and inference. Because, there is NO FACT OF SCIENCE THAT HAS EVER BEEN WRONG. Only the inferences have been wrong. Newton wasn't somehow wrong and then corrected by Einstein. Newton was entirely right within the domain of where his math applied. Einstein was not wrong and then corrected by quantum mechanics, Einstein is a FACT within a certain domain.
It's like, in a computer program,
if (a 10) use_new_theory
there's nothing WRONG with the old theory. The fact is still perfectly FACT for a 10 that is wrong.
Maybe this is not science to you or to anyone else, but this is a lot clearer of a message than any of the mishmash (oh we got our facts wrong before) gobbleygook that you lay on the public. Science is never wrong. Only inferences about known facts are.
Adhere to that message, and maybe you can stop the flood of ignorance of the bible and national enquirer from replacing really verifiable facts.
before it is too late.
This is my sig.
Adhere to that message, and maybe you can stop the flood of ignorance of the bible and national enquirer from replacing really verifiable facts.
Who said anything about the Bible?
Are you high, Todd?
You should call Alanon. They can help you.
Seriously, Todd. They can help.
Oh, and Todd? Because I do not subscribe to Slashdot, your replies will fall off bottom of my user page. If you care to reply to this message, just post a message to geomon on talk.origins. Please do start off your message with the same old and wornout routine about science as a way of introduction.
Buhbye.
"Rocky Rococo, at your cervix!"