Emil O W Kirkegaard

Won't surprise. @PsychRabble I was surprised you didn't go further in your book, but it is from 2012 before the statistical problems were as clear as they are now.

I had your 2005 review, but hadn't seen the book--just read chapter 3 and will read it all; it's fantastic. In the blog I was more careful (and have now made some updates based on what I've read so far from your book).

To be clear - I am articulating criticism of the original study (I think there is a convincing argument it is deeply flawed) and of the notion expectancy can strongly shape IQ. But I tried to distinguish (based on your review) that expectancy effects are real and important.

I do have a question. In the book you buy into Rosenthal's "low reliability makes it harder to detect significance" argument. Do you still? I'm convinced by Galman that this is not a valid defense (e.g. http://andrewgelman.com/2017/02/11/measurement-error-replication-crisis/ …)