http://www.educationnews.org/education-policy-and-politics/barry-garelick-mat...
Wow, the comments following this article are like a Math Wars All-Star Game, and both sides struggle to set up straw men fast enough to knock them back down again.But here's what really makes me mean: It's not that many of the commenters aren't trying to cite and link to research, because many of them are. But the research they should be linking to can't be linked to, because our silly academic publishing system keeps that locked away from the public eye. For example, commenter EANelson says, "For the cognitive science of learning (and teaching) math, see http://www.ams.org/notices/201010/rtx101001303p.pdf." It's a two page article with very few references, and at best it's a commentary piece, not anything close to a comprehensive take on the cognitive science of learning and teaching math. In contrast, consider a 2007 research handbook chapter written by Douglas Clements and Julie Sarama. Even after limiting themselves to just the cognitive science of learning math in early childhood, their summary of peer-reviewed research goes on for 95 very dense pages. But EANelson couldn't have linked to it, and neither (legally) can I. So while I dislike seeing the straw-man arguments, in some ways I can't blame them when the research community hasn't opened up all their results for the world to see.
(By the way, the article itself was based off think tank research, which often is available for public view and scrutiny. For my thoughts on that, see http://blog.mathed.net/2011/10/publication-paradox.html.)