More waves in longitudinal studies do not help against initial sampling error
Suppose you ran 20 studies. By coincidence, one of them attains p < .05 and, by good fortune, it's in the right direction too. The other studies get thoroughly examined…
The art of skeptical statistics
In the very enjoyable book The Art of Statistics: How to Learn from Data David Spiegelhalter mentions as example of an experiment that sounds too good to be true: In…
Educational interventions keep not working
We've been over this topic previously in 2017, but since this is the eternal idea, let's look at more data. Let's start historically. 1969 Arthur Jensen's famous article “How much…