Tuesday, July 17, 2012

I'm 95% Confident This is a Good Definition of a P Value

by Chris Masterjohn

New post over at Mother Nature Obeyed:


In this post, I explain why I consider it legitimate to read "confidence" that something is not due to chance into a P value. I discuss the differences between Bayesian and frequentist approaches, the agreement between P values derived from a significance test of the null hypothesis and those derived from the construction of a 95% confidence interval, and I use examples that include funny-looking red imps that like to steal things and lie about it.

Enjoy!

Read more about the author, Chris Masterjohn, PhD, here.