"Most scientific studies are wrong....." Topic

…says a researcher with a probable axe to grind.
link

Aren't most of the great, meaningful, civilization-making discoveries made by engineers and laymen anyway? :)

Dan

Ioannidis recommends asking the following questions: is this something that has been seen just once, or in multiple studies? Is it a small or a large study? Is this a randomized experiment? Who funded it? Are the researchers transparent?

So the concept of consilience pops up again

Would you write off an athlete from competing a Marathon because they made a spelling mistake on their application form ?

Publishing a paper is only one of the steps in the system and is the greatest source of misconceptions.

Getting a paper published is no great feat, people have submitted random text and passed muster and it is like applying to the Marathon, there is no guarantee you'll even be able to cover the distance.

It has to be peer-reviewed and be found to be consistently right whenever it is tested before you can really make a valid statement.

Science is incredibly boring. People are going blind staring at excel sheets and old IBM greenscreen monitors trying to sift data into meaningful information. And if your grad student had a bad day your incredible discovery is the result of a few mistyped digits.

I also want to point out "reader's bias" who already comes predisposed to assume that the vast majority of people doing such work are either incompetent or are hell-bent on releasing false information. This kind of article is a godsend because the title alone will reinforce beliefs and no need to read the article : "it's proven, they're all frauds !"

What the article is saying is that people get sloppy, make mistakes, use the wrong methods, their statistics skills are faulty etc …

And the media is also to blame in this instance. The propensity to sift through every single paper published or submitted for review to systematically pick the weirdest ones or to create sensationalist headlines that will trick people into reading a barely edited piece of standard block text which usually reads like "Scientists have discovered that [insert random object or substance here] will give you cancer. Yadda-Yadda

And when you do bother to check the paper it's usually a test where it turns out that if you overdose on the stuff for decades on end your risk of cancer is correlated to go up with a few % And unless we are talking about really potent chemicals, that's pretty much 99% of such articles explained.

There are frauds and you know what ? They get constantly exposed ! We know they are frauds because they are found out, but somehow some people want to believe that fraud can be kept out of sight and hundreds of thousands of people involved will either not notice anything or be an active participant and none of the parties that might suffer major damage ever seem to even worry about all those conspiracies everyone seems to be so keenly aware of.

Take all the medical conspiracies people on this forum can cite. How come private insurance, the people who will find a way to deny you paying for a stupid band-aid are more than happy to defrauded at the tune of billions of dollars by "Big Pharma", people have been thrown into the East River with concrete galoshes for less than that …

Aren't most of the great, meaningful, civilization-making discoveries made by engineers and laymen anyway?

No.

And that goes with the understanding that engineers and laymen go back millennia, while scientists are a relatively recent calling.

"Aren't most of the great, meaningful, civilization-making discoveries made by engineers and laymen anyway?"

Wouldn't that depend upon the definition of 'great,
meaningful, civilization-making discoveries' ?

Not that engineers/laymen were responsible, but there's
a decent argument against the bellicose use of nuclear
power as 'civilization-making' and the same might be
true of the discovery that twisted sinew could be used
to toss sharpened spears farther than a human arm
or the fact that great levers activated with heavy
weights could toss very heavy stones at folks or
walls.

These latter falling more into the province of engineers
and laymen than in that of scientists.

I can believe this.

Most of the individuals doing the study have something that they are trying to prove or a Agenda to keep.

So they end up using only data that proves what they are trying to prove.

Most of the individuals doing the study have something that they are trying to prove or a Agenda to keep.

How would you "know" this? How do you "know" it's "most of the individuals"? Are you just making stuff up again?

I can believe this.

Most of the individuals doing the study have something that they are trying to prove or a Agenda to keep.

So they end up using only data that proves what they are trying to prove.

The projection is strong in this one

I seem to have fallen into an endless loop here.
Or maybe a Möbius strip.
The guys who put out a study accusing "most" studies of cherry picking are accused of cherry picking.

The guys who put out a study accusing "most" studies of cherry picking are accused of cherry picking.

Where, on this thread, did you read this? I've read to original posting and nowhere does Prof Ioannidis mention anything about cherry picking. He does make other claims, like small sample sizes and a limited statistical understanding and even sloppiness. But I think your characterization is not what Ioannidis actually says.

Mithmee?

I've become immune to Mithmeeisms, I guess.