Talk:Psych Science - June 2018/@comment-35453425-20180626031730

The Pratt study sounds really good, and I think I like the design. Neat way to tease apart two hypotheses/models. I do not know a lot about this area, though, so maybe others have thoughts?

I knew about Joshi and Fast (2013), and I never believed that. I think studies like that are really what have gotten us sidelined in the past 15 years in psych science, and have directly created the "nothing replicates in psychology" movement. Which is really building. We have to stop taking well-studied phenomena (like temporal discounting) and "shot gunning" other phenomena (i.e., cute things like "power") at them to see what sticks, then running off. I am not opposed to the idea of studying how having power makes one behave/think differently. In fact, I think it is really important. But, if you get this kind of result, don't stop there, see if it holds for other things that ALSO hold with temporal discounting. See if power (or whatever else you want to add) is consistent in how it interacts with all of the usual (well-studied) players in a psychological relationship of interest.

I know many of you are sick of hearing this, but pre-registration will save us in the future. In two ways (I hope!): 1) More science done well, and reported no matter the outcome, thereby making what are now sexy one-off findings much more apparent as exceptions to the rule, and 2) a cleaner way to evaluate academic progress and accomplishment, where pubs still matter, but so too does having good ideas that might be true, but turn out not to be.

Steps off high horse.