Over half of psychology studies fail reproducibility test

Approximate replication rates in psychology: social 37% cognitive 42% personality 55% clinical 44% So a list of famous psychology experiments that do replicate may be shorter. https://www.nature.com/articles/nature.2015.18248

> Can you link a study, that makes such a claim? Of course I can, I took it from a study! https://www.nature.com/articles/nature.2015.18248 -------------------- some other studies on the subject, not all agreeing on the conclusions, as it is natural https://pubmed.ncbi.nlm.nih.gov/29861517/ https://www.sciencealert.com/two-more-classic-psychology-stu... https://www.pnas.org/doi/10.1073/pnas.2208863120 the real problem is that the kind of psychology studies that gets the more attention on the media, that shape the culture and the beliefs of the general public and that are often used by governs and institutions (even medical ones) to deploy policies, health care plans, school or work environment guidelines, etc. etc. are those that are less trustworthy, that fail to be replicated the most and more subject to interpretations and subjective biases, in particular social psychology (citation: Social Psychology (38% success rate, n = 256).)

The title of this study "No link found between psychedelics and psychosis" should really be that "the authors did not find links between psychedelics and psychosis using their survey method". These survey methods come with huge uncertainty and lots of pitfalls. I would not be surprised if the result in the paper is not reproducible. In fact, "over half of psychology studies fail reproducibility test"[1] [1] https://www.nature.com/articles/nature.2015.18248

Just for this argument: "Research shows that as many as 75 percent of personality test-takers receive a different result when tested again." There is an easy counter: https://www.nature.com/articles/nature.2015.18248 The other two still holds for me though. But Big Five is not any better IMHO, it can be a trait for psychological health, but definitely not for personality.

The more surprising part to me was that quote about the 2015 study that tried to reproduce 100 studies and managed to replicate 39. That's honestly way better than I expected for psychology. Here is the paper: https://www.nature.com/articles/nature.2015.18248

Reminder that over half of psychology studies fail reproducibility test https://www.nature.com/articles/nature.2015.18248 https://statmodeling.stat.columbia.edu/2016/03/03/more-on-re... I'm not an expert in psychology but something about that research just feels off. Seems like a real hard thing to quantify and detailed year over year results like they're presenting should be hard to get.

Yes, a significant proportion of all scientific research is not reproducible e.g., "Over half of psychology studies fail reproducibility test" https://www.nature.com/articles/nature.2015.18248 Cancer research: "scientific findings were confirmed in only 6 (11%) cases." https://www.nature.com/articles/483531a

I tend to be skeptical of most studies that come out of psychology. Over half of them seem to be junk - https://www.nature.com/articles/nature.2015.18248 That's not to say other areas are better by default. There is just a lot of junk science out there. "Study says X" is meaningless to me personally. "Study X was reproduced by A, B, C" is something I'd take notice of.