Hot take: Poor quality research is effective because of the imbalance in the perceived value of theory versus evidence. What people really like are explanations or theories ("understanding" something makes you feel smart). However, without compelling evidence, any explanation is just a story.
Fully agreed, ideas are somewhat cheap, even good ones. The hard part is always doing the work. Hence, why I am never really concerned about being scooped. If someone else is willing to do the work, good for them and I guess less work for me.
I love this video; I really cannot believe that the dots do not change their size. I noticed however that the orange dots move. Is this necessary to get this strong effect or could you get a similarly strong effect with orange dots that stay in place and only the surrounding dots move?
The unrelated cherry on top is that the NHB paper arguing that preregistration leads to high replicability was apparently retracted. Presumably because their preregistration didn't match the reported analysis. bsky.app/profile/jess...
In other science reform implosion news, the contested article including OSF & Data Colada authors on how preregistration & other rigor-enhanching practices led to high observed replicability (the one that couldn't produce its own preregistration) was just retracted www.nature.com/articles/s41...
Four labs discovered and replicated 16 novel findings with practices such as preregistration, large sample sizes and replication fidelity. Their findings suggest that with best practices, high replica...
I have the feeling correlation might give you a better idea than this graph. It looks quite similar, but you do not know how often the order of values changes.
Barchard, K. A. (2012). Examining the reliability of interval level data using root mean square differences and concordance correlation coefficients. Psychological Methods, 17(2), 294–308. doi.org/10.1037/a002...web.archive.org/web/20170810...
How about simple correlation or concordance correlation coefficient? If the value is high enough, both variables are sufficiently similar.
Congrats to starting your new lab Julia! What a loss for Amsterdam and win for Potsdam. 🥳