Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

A Big Study About Honesty Turns Out to Be Based on Fake Data

buzzfeednews.com

Post reply on HN
Print view View original

71–80 of 98 posts

Re: A Big Study About Honesty Turns Out to Be Based on Fake Data

#71

Post by WalterBright »

iratewizard wrote: post #58
BurningFrog wrote: post #43

Earlier quoted context omitted.

Especially when it's only a single study with a surprising result. Those go viral easily, but are usually wrong.

The replication crisis has been a long standing problem in social sciences and medicine. The former's darkest stain is its collective need to prove their preconceived world view. The latter's seems to be motivated by profit. I'm sure nobody is surprised that zealotry and greed have made some of science as reliable as psychic readings.

> The latter's seems to be motivated by profit.

The same problem persists in communist countries that don't allow profit.

Re: A Big Study About Honesty Turns Out to Be Based on Fake Data

#72

Post by azinman2 »

WalterBright wrote: post #37
vegetablepotpie wrote: post #20

Earlier quoted context omitted.

> I’m sick and tired of people saying “the science is settled” or “trust the science”. You’re right, but also, there’s more to it. There is a difference between what people say and what people mean and context is important. When I’ve heard people say “trust the science” it is always directed to people who have demonstrated that they do not understand the scientific method and the meaning of the word “healthy skeptici…

When I hear people say "trust the science" the speaker is often a non-scientist who is selectively picking what science fits their agenda and ignoring the rest. It's the same thing when somebody says "trust me, I'm a XXXXX", it's time to be very skeptical.

The problem is, not all opinions are equal. And the internet, including forums like this, have no notion of status, authority, or stature — so it puts all opinions, no matter how informed they are, on the same page. And while status and authority doesn’t mean one is right, it is important than on average our information sources are more informed than the average person who has no training, no relevant knowledge, no domain expertise, and little ability to disambiguate between contradictory or even harmful information. This tendency has spilled over into the nearly infinite sources of media and information that now exist, with accelerants that didn’t previously exist for bad information to travel wide, far, and fast.

So when people say trust the science, what they’re saying is a bunch of people who do this all day every day are collectively hive minding some opinion on how things work. If it’s wrong, it will be wrong for likely non-obvious reasons. And thus that’s the best we as humans can hope for at any point in time. As a race we are continually learning and updating our understanding of the world; just because it’s not perfect at any one time doesn’t mean everyone should just throw away all that they know and assume everything is wrong and up for grabs.

Re: A Big Study About Honesty Turns Out to Be Based on Fake Data

#73

Post by jollybean »

efitz wrote: post #7

Science is a process, a way to find truth. It is not in itself truth. It is subject to all the same human flaws and venality as any other human endeavor. Just because something is published in a paper doesn’t make it true. Just because the paper was peer reviewed doesn’t make it true. Just because there is a “consensus” doesn’t make it true. I’m sick and tired of people saying “the science is settled” or “trust the s…

"Science is a process, a way to find truth. It is not in itself truth."

We mostly all know this.

The issue is populism and communications.

When we use 'Science' as a basis of infallible credibility in some areas, then it's understandable that some are 'shocked' when that infallibility is obviously not true, and then people become jaded and lose confidence.

There's no language to differentiate between the murky greyness of some things (i.e. memory and recall), and the relatively unambiguous results of other bits of research (i.e. acetaminophen is safe).

Why should the non-scientific public, even sometimes, have confidence in a system that is so often very wrong? How are they supposed to know which bits are 'effectively true' and which are 'grey'?

From the outside, it looks like Science is being simultaneously authoritative (to the point of moralizing) and in other areas totally wrong but lacking in self awareness while saying 'Oh, that was wrong, bu Science is a process of discovery, it's not always right' etc..

Given that Science is constantly 'changing it's mind' it's perfectly reasonable for regular people to doubt Climate Change science: the 'authorities are often wrong, ergo, they may very well be wrong here'.

If Science doesn't have a way of effectively (and by that I mean simply and clearly) communicating the degree of confidence in something, then we're playing a very dangerous game with credibility of the institution.

Re: A Big Study About Honesty Turns Out to Be Based on Fake Data

#74

Post by hex4def6 »

WalterBright wrote: post #71
iratewizard wrote: post #58

Earlier quoted context omitted.

The replication crisis has been a long standing problem in social sciences and medicine. The former's darkest stain is its collective need to prove their preconceived world view. The latter's seems to be motivated by profit. I'm sure nobody is surprised that zealotry and greed have made some of science as reliable as psychic readings.

> The latter's seems to be motivated by profit. The same problem persists in communist countries that don't allow profit.

I think you can replace profit with "power". Publish or perish is true in most academic circles, regardless of government.

Re: A Big Study About Honesty Turns Out to Be Based on Fake Data

#75

Post by jollybean »

vegetablepotpie wrote: post #20
efitz wrote: post #7

Science is a process, a way to find truth. It is not in itself truth. It is subject to all the same human flaws and venality as any other human endeavor. Just because something is published in a paper doesn’t make it true. Just because the paper was peer reviewed doesn’t make it true. Just because there is a “consensus” doesn’t make it true. I’m sick and tired of people saying “the science is settled” or “trust the s…

> I’m sick and tired of people saying “the science is settled” or “trust the science”. You’re right, but also, there’s more to it. There is a difference between what people say and what people mean and context is important. When I’ve heard people say “trust the science” it is always directed to people who have demonstrated that they do not understand the scientific method and the meaning of the word “healthy skeptici…

"Certainly it would be better to have an informed and well educated citizenry. "

Of course it would, but there's no way for people to 'stay informed' on the multitude of various things.

The system we live in is fundamentally based on legitimate authority. We have no reasonable way to try to debate with our doctors, dentists, lawyers, engineers. It's literally why they exist - to understand, internalize and work with the inherent truth in a system.

We trust them, that's the way it works.

So how does a system that is inherently grey, be 'wrong a lot of time' but 'right in others' in a manner that is consequential (i.e. vaccines, climate change).

That's a tough social problem.

Re: A Big Study About Honesty Turns Out to Be Based on Fake Data

#76

Post by azinman2 »

> Looking at the data only when they were aggregated, anonymized, and sent to him, he said, freed him from the work of securing ethics approval from the university to perform research on human subjects.

Note he was kicked out of MIT for doing work not approved by the IRB if I recall correctly.

Re: A Big Study About Honesty Turns Out to Be Based on Fake Data

#77

Post by spookthesunset »

efitz wrote: post #7

Science is a process, a way to find truth. It is not in itself truth. It is subject to all the same human flaws and venality as any other human endeavor. Just because something is published in a paper doesn’t make it true. Just because the paper was peer reviewed doesn’t make it true. Just because there is a “consensus” doesn’t make it true. I’m sick and tired of people saying “the science is settled” or “trust the s…

> I’m sick and tired of people saying “the science is settled” or “trust the science”.

Or "I follow the science". Nope, that's not how this works. Science doesn't tell people what to do. It isn't a book of instructions or a fixed set of truths. It's a process to learn how things work. What you do with those learnings is a squishy human meatspace thing involving values, politics, social norms, traditions, emotions, etc.

Anybody who says they "follow the science" is not doing any such thing.

Re: A Big Study About Honesty Turns Out to Be Based on Fake Data

#78

Post by throwawayboise »

enkid wrote: post #32
throwawayboise wrote: post #29

Earlier quoted context omitted.

> When I’ve heard people say “trust the science” it is always directed to people who have demonstrated that they do not understand the scientific method I think "always" is too strong. With regards to COVID vaccines I have heard that line thrown at people who say "this conclusion is too new, it has not stood the test of time" and now lo and behold we are seeing that the early "95% effective" rates are not holding up.…

The 95% effectiveness was based on the version of COVID that was circulating at the time. No one said it would be 95% effective against every possible mutation. Claiming the results of those studies are wrong instead of a result of a changing environment is itself a misunderstanding of science.

Claiming that the vaccines were necessary to "end the pandemic" without factoring in the possibility of variants for which the vaccines were not as effective was not scientific thinking.

Re: A Big Study About Honesty Turns Out to Be Based on Fake Data

#79

Post by Falandafa2021 »

azinman2 wrote: post #72
WalterBright wrote: post #37

Earlier quoted context omitted.

When I hear people say "trust the science" the speaker is often a non-scientist who is selectively picking what science fits their agenda and ignoring the rest. It's the same thing when somebody says "trust me, I'm a XXXXX", it's time to be very skeptical.

The problem is, not all opinions are equal. And the internet, including forums like this, have no notion of status, authority, or stature — so it puts all opinions, no matter how informed they are, on the same page. And while status and authority doesn’t mean one is right, it is important than on average our information sources are more informed than the average person who has no training, no relevant knowledge, no d…

> So when people say trust the science, what they’re saying is a bunch of people who do this all day every day are collectively hive minding some opinion on how things work.

Only the right people who do this every day though... Even better only the ones whose reputation and parts of their salary depend upon more people believing in their science and not questioning them.

Re: A Big Study About Honesty Turns Out to Be Based on Fake Data

#80

Post by a-nikolaev »

rozab wrote: post #36
koboll wrote: post #10

"Now some are questioning whether the scientist himself is being dishonest" kinda undersells it. The debunking offers numerous pieces of stone cold proof that the paper is straight-up academic fraud: https://datacolada.org/98 And since the post says "the fourth author has made it clear to us that he was the only author in touch with the insurance company", it seems clear that Ariely personally fabricated the data.

This evidence is clearly irrefutable. It's fascinating how a well respected scientist could make such elementary mistakes when fabricating data.

It's easier to gain reputation if you are fine with cutting corners here and there, or even make up things as long as no one can see that. And once you are at Harvard (or Duke for that matter), most people wouldn't even question your credibility.

EDIT: And to the point of not being able to fake that data well. Yeah, again, if we are in the business of getting credit points quickly, faking the data quickly makes sense too. No one would take a close look, right?

Post reply on HN