Here’s my excuse for using obsolete, sub-optimal, or inadequate statistical methods or using a method irresponsibly.

E. J. Wagenmakers writes:

My colleague Klaas Sijtsma wrote a book titled “Never waste a good crisis” and I [E. J.] am reviewing it here.

His main claim is that questionable research practices would be much reduced if researchers would seek advice or collaborate with professional statisticians or methodologists. I thought it was refreshing to see it put so clearly. Maybe this is a claim that can be empirically tested. Anyhow, you might have an opinion on it and I thought it could be something for your blog.

The review is interesting, and the reviewed book looks interesting too.

But I disagree with this quoted statement from the book: “There is little if any excuse for using obsolete, sub-optimal, or inadequate statistical methods or using a method irresponsibly.”

Don’t forget about the concept of bounded rationality! It can be costly for an applied researcher, or even a statistical expert, to tool up and learn how to correctly use up-to-date and optimal methods (even if we accept the generally dubious idea that there is in general an “optimal” statistical method for any given applied problem).