When your field can only have less than half of its top papers reproduced in full, something is very wrong with your field. It's to the point where one can safely assume that any psych paper (but especially social psych) is garbage until such time as it's been replicated a time or two.
[1]: https://www.nature.com/news/over-half-of-psychology-studies-...
Good, because it is probable that what we should assume this for any science[1].
[1]: http://journals.plos.org/plosmedicine/article?id=10.1371/jou...
In reality, the current scientific culture is one of rampant distractions, where pretty much everyone is running around like a beheaded chicken. Woe unto those that might point out that a prominent researcher with tenure isn't using sound methods. That, as well as a generally toxic culture that makes Wall Street look good, was part of why I left.
There have been some efforts towards creating a better culture (in particular, I've had my eye on Popper [1] and the "SciOps" movement and Software Carpentry [2], as well as various open access efforts). At the end of the day though, scarcity in terms of funding plus lack of job security plus medieval power dynamics are holding any substantial cultural shift back.
[1] http://falsifiable.us/ [2] https://software-carpentry.org/
[0] https://en.wikipedia.org/wiki/Replication_crisis#Overall
What that says is that e.g. 69% of scientists working in physics and engineering had failed to replicate some experiment at least once in their career. In other words if they performed 1000 replication attempts of other scientists' experiments and failed to reproduce the same results once, they would be counted in that survey. By contrast in psychology, one hallmark experiment tried to replicate the findings of 100 key studies in psychology from highly regarded psychology journals. And it was unable to replicate the findings of 64% of the studies. What's that number for physics? No idea other than that it's going to be much lower.
Make sense? Wiki is, almost as a rule, just completely awful for contemporary or potentially controversial topics and this is a textbook example of that.
I think this speaks more to the nature of academia, and the commercialisation of publication than the validity of the field in and of itself.
I mean, just because it's in a psychology textbook doesn't necessarily mean that's it's BS... But that's not a decent standard. Even astrology clears that bar.
More recent critical analysis has poked a lot of holes in Milgram [0].
Important work no doubt but not quite the smoking gun it has been proclaimed to be.
[0] https://www.theatlantic.com/health/archive/2015/01/rethinkin...
> It’s a phenomenon that’s been used to explain atrocities from the Holocaust to the Vietnam War’s My Lai massacre to the abuse of prisoners at Abu Ghraib.
This sounds an awful like the sentence in the Lifespan of a Lie:
> It has been invoked to explain the massacre at My Lai during the Vietnam War, the Armenian genocide, and the horrors of the Holocaust.
We seem to be using a lot of these experiments to explain the same things ...
It seems to me that a lot of the seminal social psychology studies happen around this time ...