Bruno: The *Real* Causes Of "Teaching To The Test"
I've made previously: that many concerns about "teaching to the test" are at least partially misguided.
As he points out, much of the marginalization of science in elementary schools predates NCLB, which suggests that curriculum narrowing can't be entirely explained by high-stakes testing. (A more likely culprit? Only 1/3 of elementary school teachers feel prepared to teach science in the first place.)
Additionally, Willingham elaborates on the importance of teaching content to promote reading comprehension. Even as late as 3rd grade elementary students are spending nearly half of their school time on English Language Arts, which leaves little time for the numerous other subjects - like science and history - that are so important to building students' content knowledge and, in turn, their ability to understand what they read.
To reiterate, I think it's pretty clear that many of the practices that are labeled "teaching to the test" are, in fact, problematic. What I believe we need to take more seriously, however, is the possibility that these practices are attributable as much to other factors (e.g., misconceptions about educational psychology and inadequate teacher preparation) as they are to the perverse incentives of high-stakes testing. If we're piling all of the blame on NCLB we're probably misdiagnosing the problem. - PB (@MrPABruno)