Hacker News new | past | comments | ask | show | jobs | submit login

The problem with social sciences isn’t that they’re young, it’s that they’re all junk science. They’re all filled with poorly defined experiments, theories that you can never properly test, results that you cannot reproduce and that nobody’s going to try to, and if the topic of a study is even slightly political or controversial, you’ll often find that a study can’t feasibly be conducted, or it will be conducted by people who don’t want it to be rigorous in the first place.

I’m skeptical that a lot of these questions even can be answered scientifically at all, but rather confident that they’re not going to be by the existing system.




There is plenty of junk, I'll grant you that. But there is also variation in study quality, which ipso facto implies that some studies might just even be good. I take what you are talking about to be the relative immaturity of social sciences.


> But there is also variation in study quality, which ipso facto implies that some studies might just even be good.

The fact that there's variation doesn't at all imply that any of it is any good. It could vary from "embarrassingly flat-earth-theory-in-2024" bad, to "unable to produce a falsifiable hypothesis" bad.

And even when some of it is good (controls, reproducability, etc), it gets completely ignored by practitioners[1]. Imagine if doctors prescribed eye of newt even when there's studies proving the efficacy of paracetemol.

[1] By practitioners I mean therapy and therapists. You could send 100 random people to 100 different therapists and all of them would report that a followup visit has been recommended. As far as therapists are concerned, there is no such thing as "You're perfectly normal. Congratulations and come back only when you have a problem"




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

Search: