Hacker News new | ask | show | jobs
by native_samples 1712 days ago
The original claim doesn't cite any peer reviewed research.

Nonetheless, there are enough really bad peer reviewed studies out there (which get attention) that it's not insane to willingly assume the worst.

1 comments

I'd just like some acknowledgement that maybe - just maybe! - confirmation bias is kicking in when people make these assumptions.

Maybe the reason for this particular brand of reflexive skepticism is not genuine concern about the quality of the research, but an attempt to rationalize their own actions and beliefs without actually doing any introspection or self-examination.

Flip it around: why aren't you the one engaging in confirmation bias?

Look at the comment down-thread. Only 44% of a set of highly cited clinical trials had been successfully replicated and 32% appeared to be wrong when replications were done. That's for clinical trials, let alone stuff like social psych! Given a random clinical trial it is in fact pretty reasonable to assume it's wrong, and might even be rational, given that you'd hope highly cited and impactful studies are more likely to be correct than not.

But don't take my word for it. Take it from a former editor of the BMJ:

https://blogs.bmj.com/bmj/2021/07/05/time-to-assume-that-hea...

"Science" really is just that terrible, unfortunately.

44% replicability is a lot higher than the average right wing internet meme/argument about gun violence and gun safety.