How can you know if it’s *really* “research-based?”

Daniel Willingham:

My new book, When Can You Trust the Experts: How to Tell Good Science from Bad in Education is now available. (There’s a link for a free download of Chapter 1 on this page.) EDIT (7:17 pm 7/18) I just saw that it’s listed as “Recommended,” with a micro-review on the Scientific American website.
I wrote the book out of frustration with a particular problem: the word “research” has become meaningless in education. Every product is claimed to be research-based. But we all know that can’t be the case. How are teachers and administrators supposed to know which claims are valid?
It’s notable that this problem exists in many other fields. However good your training, research doesn’t stand still. So how does a pediatrician who has been in practice 10 years know that what she learned in medical school as the optimal treatment for, say, croup, is still the best treatment?