4 ms·
Another victim in the "replication crisis" bloodbath for psychology. Personally, I'm probably going to wait 5 years before I believe any psychology science cla
by BurningFrog 7y ago
Another victim in the "replication crisis" bloodbath for psychology.
Personally, I'm probably going to wait 5 years before I believe any psychology science claim. Especially if it confirms something I already believe in :)
- mensetmanusman 7y agoYes, this has been a disaster that will be written about in the history books. I hope they take down all the TED talks about these papers that have been debunked. In hindsight, it is kind of funny in a dark-humor way.
- username90 7y agoTestosterone increases blood flow, it could be that the participants taking it got more alert meaning better scores but that testosterone also worsens results for similar alertness levels.
- Jolter 7y agoThat's just not what the study studied. You can speculate but there is no evidence for that hypothesis in this paper.
- username90 7y agoI didn't say that this study proved it, I just meant that effects like that could cause the non-linear results reported and mess with studies.
- collyw 7y agoWhat is the rate of failure for "harder" sciences, does anyone know? In saying that I think there will be a self correcting mechanism in more physically based science. If something can be replicated it can be built upon and new discoveries made. If something can't be replicated it will end up being forgotten about. Also I do think that science needs a way of reporting failed studies and failed replication. At the moment all the glory and funding go to studies that produce positive results. Producing a failed result is still valid science and should be doccumented.
- bumby 7y agoThere has been some attempt at this, with the SURE journal for unsurpising economics outcomes being one example. Unfortunately, this approach seems to be at odds with the incentive structure in academia
- BeetleB 7y agoIn all my time in academia, in my discipline (cross between engineering and physics), I never heard anyone try to replicate a study. No advisor would allow students to "waste" time on it. They were also pretty open in vocalizing that they didn't believe the result in paper X in journal Y. They knew the problem existed.
- BurningFrog 7y agoPretty sure dead and unconscious things are way easier to do objective science on. Psychology must be one of the hardest disciplines to design experiments for.
- gerbilly 7y agoIt's a kind HN received wisdom that all psychology studies are false or invalid. Personally, I see this blanket dismissal as one of the blind spots of our otherwise intelligent community. Perhaps because so many of us are engineers, and that it strokes our egos to see the softer sciences fail?
- Fr0styMatt88 7y agoPsychology is like trying to debug a program with nearly zero tools, not even a reliable ‘printf(..)’. I don’t think it’s ego-stroking so much as it is just frustration at the seemingly intractable complexity of human behaviour.
- ModernMech 7y ago> trying to debug a program with nearly zero tools, not even a reliable ‘printf(..)’. That's called kernel development (from scratch) :P
- BurningFrog 7y agoThere are surely many valid ones. Problem is, right now you can't tell which they are.
- naasking 7y ago> It's a kind HN received wisdom that all psychology studies are false or invalid. Not ALL, but the replication crisis has demonstrated that at least 2/3 of such studies can't be replicated. If you had a communication channel where 2/3 of your packets failed to deliver, I'd say a statement like "this channel isn't reliable" is very justified.
- merpnderp 7y agoThat's 2/3rd of medical research paper where the standards are the highest. I'd be shocked if the rate was only a mere 66% everywhere else. You'd have to be a fool to take on faith anything published today that wasn't replicated, hopefully more than once.
- evrydayhustling 7y agoThen, wait another 15+ years before that claim stops being widely held and repeated as a justification for policy, business decisions, etc.
- justnotworthit 7y agoWait, you did or you didn't believe this paper's psychology science claim? Why or why not? If we/you apply your criteria to other findings, will we find the good science?