Hacker Newsnew | past | comments | ask | show | jobs | submitlogin

I'm a bit of a contrarian, as some of you may know. When the crowd goes one way, I look to see what the other way looks like.

I find articles like this -- where the underlying premise seems to be "people are stupid, science is not" to be deeply flawed. People use the word "science" to cover all sorts of things. In this example, it was used to cover the results of a study. Guys? Studies are inherently non-conclusive -- for lots of reasons, including the correlation and causation problem. I can show you studies that will show correlations between about any two things you would like. What's the old saying? "If you torture the data enough, you can get it to confess to anything". The stats on medical studies, in particular, are very concerning. The proper response to being presented with any study is to say something like "That's interesting. What other pieces of data do we have?"

It's not to change your world-view. If your world-view is so flimsy that you'll change it on a dime, you're not exactly being a critical consumer of information.

Secondly scientists are people too, just like the ones studied.. Everybody seems to overlook this fact. We all have these wonderful gems of how people act irrationally, self-reinforce in groups, are slow to change opinions, etc., and nobody asks "wonder how all of that affects the study of science?"

Why? Because the purpose of such self-congratulatory bullshit like this is to tell ourselves "I am a creature of reason and science. These other people are primitive non-thinking dweebs"

It's all just so much intellectual self-stimulation.

I'm all for studies in human irrationality to continue. This is a very interesting and fruitful area of research. But the picking it apart and trying to make social or political observations out of it? Not so much.



Secondly scientists are people too, just like the ones studied.. Everybody seems to overlook this fact.

I disagree that people overlook that fact. Humans are certainly not infallible, and that's why peer review is an important part of the process for confirming scientific discoveries and claims. That's what is so great about science; claimed results must be reproducible independent of the scientist, otherwise the claims are generally discarded.


> I disagree that people overlook that fact. Humans are certainly not infallible, and that's why peer review is an important part of the process for confirming scientific discoveries and claims.

Peer review compares to the literature and does some minor arithmetic checks. It rarely redoes the statistics, let alone verify that the measurements were real.

That's part of why peer review is almost useless in detecting fraud. It has other failings as well.

It is better than nothing.


I agree, but I meant peer review in an objective sense. There is nothing stopping anyone else from verifying their own measurements. The bigger and more unbelievable the claim, the more subject it is to scrutiny.




Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: