xTYBALTx

From the WSJ:

__________________
Current good song: Justice - Stress

Beatnik

There's something ironic about a wildly misleading thread headline on the topic of misleading scientific analysis.

Just sayin.

__________________

P8riot

But what do you think of the article?

__________________





Age of Conan: Tharashk (Taxiarchos), Thaumiel (alt), guild Ars Tactika - Cimmeria

TheFeniX

Quoted for truth. I mean, how many tries do you think it took them to get WD-40 right?

__________________

Beatnik

39?

__________________

leejo

No it isn't! If error is as much a part of science as discovery, then I'm Einstein.

For science to be effective, discovery needs to be a much bigger part than error. Otherwise we're just randomly blundering along. QFT? Scientists of all people should be taking greater care to ensure their methods and calculations are sound.

39 is an awesome guess, though, Beatnik. Kudos.

AMosely

I do and do not get the point of this opinion piece, which has even less credence than most modern scientific studies. I can understand that far too many scientists are primarily aimed at securing their next grant, but on the other hand, what do you expect from a second-guessed, privately funded special interest-based scientific community? The absolute truth? Do we expect the same of judges and doctors? Pastors? How about Presidents as they march a country to war?

Science is trial and error. The alternative is religious belief. At least with scientific study we stand to learn something through the collision of photons, molecular biology, and the exploration of the universe. At least with science the guidelines it offers are based on something other than century-old superstition.

In a world full of baloney, scientific study is the closest thing we have to real meat. It shouldn't surprise anyone that often times you just run into more baloney. Just keep digging.

ednos

If you think science is bologna, you should look into politics.

__________________

The trouble with the world is that the stupid are cocksure and the intelligent are full of doubt. ~ Bertrand Russell
I have a tendency to key out three or four things and then let them battle for supremacy while I key, so there's a lot of backspacing as potential statements are slaughtered and eaten by the victors. ~ Magna Centipede

VegaBond

Was just about to type that when I saw this thread... lol

__________________

D0ubl3Hel1x

For some reason I thought this was going to be a creation vs evolution thread. I'm glad I was wrong.

__________________

sordavie

In one sense this guy is trivially right. In another sense, if 'truth' is taken loosely, what this guy says isn't surprising nor highly problematic. Then again, scientists are notoriously terrible at doing philosophy of science--the study of the nature or structure of science and scientific methodology. The problem is that scientific methodologies don't apply to theories about the nature or structure of science and scientists are trained to do science and not philosophy of science.

The naive view, which we teach high schoolers and undergrads, that science progresses by observation, hypothesis, experimentation, confirmation/revision of hypothesis is just wrong. The central goal of science is problem solving, not finding truth. So, it's no surprise that most, if not all scientific theories are, strictly speaking, false. They are only better or worse depending on what problems the scientific community looks to solve. This may require approximating truth to some more or less degree, but that's not necessary. On the other hand, if we're talking loosely about "truth," what I take it he means is that lots of scientists are sloppy with their results. That they aren't as accurate as they should be. But this isn't surprising, nor is it a problem so long as they are accurate enough that their work contributes to solving what problems scientists want to solve.

The basic argument against the naive view is that scientific theories are underdetermined by observational evidence. That is to say that any particular observation or series of observations is compatible with many incompatible theories. So, one can hardly say that these observations verify this particular theory over any rival theories that are also consistent with the observational evidence. If this is so, then the naive model of science is just wrong because the naive view is that observational evidence justifies one theory over the rivals.

If you're interested in this, Thomas Kuhn's "The Structure of Scientific Revolutions" is one of the first to attack the naive view.

__________________

Val

I LoLed till I cried.

xTYBALTx

Guess I'm the only person who was surprised to see that, according to one peer reviewed and published study, over 50% of peer reviewed and published studies contain significant errors.

I don't see how it's naive to expect better accuracy than a coin flip from peer reviewed and published scientific studies.

__________________
Current good song: Justice - Stress

sordavie

The point is that there is no such thing as accuracy. The mistaken assumption of this article's study is that the naive model of science is the correct one, and that science strives for truth. Given that, it may appear that his findings are very surprising or problematic for science. But, drop that assumption and it's not surprising or problematic at all.

This guy appears to be completely ignorant of any 20th century philosophy of science. There has always been concern that the majority of research findings are false. This is not a new idea. If you read the abstract to the article, it appears he's got some very primitive conceptions of the problems that philosophers of science have discussed for 80+ years. The "solution" is to give up the naive conception of science and see it for what it really is. And that the majority of scientific findings are strictly false is not abnormal.

Medical journals are probably not the best place to find good peer reviewed articles on topics like philosophy of science or medical ethics. They usually contain very subpar articles written by scientists or even doctors who have little or no training in those fields. Many of these aren't even peer reviewed. Why they allow this is beyond me. Good philosophy journals don't publish physics articles written by philosophers who have no training in physics. And, if you do find one, it's probably best to just consider it junk and go find a real physicist who has something to say about physics.

Here, this guy is raising concerns about scientific methodology. But that's not a scientific topic. That's a philosophy of science topic. Silly Epidemiologist.

__________________

Steeler

Not helping matters any further, in addition to the normal ossification of opinion as publications age and their editorial staff becomes more concerned with prestige than inquiry, there are now a plethora of "scientific journals" published with the sole intent of issue advocacy.

Combine that with an incautious wider media structure that values headlines over content, and you get a phenomenal amount of bad "information" injected into the public discourse.

Are eggs good for you?

__________________