Every assessment choice is a value judgment

Howard Gardener replied to a NYT article "Nonfiction Curriculum Enhanced Reading Skills, Study Finds" (published March 12, 2012) with this:

It is instructive to know that second graders who received a Core Knowledge curriculum performed better tahn comparison gropus on measures of reading. But every choice of curriculum -- and, more important, every choice of an assessment measure -- entails a value judgment. Those educators who selected a reading program that valued fictional works (NOT those who chose the Core Knowledge Curriculum in the study) presumably thought that was an appropriate emphasis. It is now up to those educators to provide measures that might reveal better performances on their currciulum -- for example, richer imagination by students or a greater liklihood of reading books of any sort outside the school environment.

Today I was part of a discussion about rearranging curriculum so that teachers can measure student growth in writing. One teacher, Jim, noted that one unintended side-effect of this kind of thinking could make him shy away from asking kids to engage in complex, messy writing assessments which might blow up in his face, but might also yield some really good thinking and writing. Jim said, "I would be tempted to stick to the things that I know would show results." When you build a rubric, or an assessment, you're always drawing boundaries around what's important, what's worth looking at in student writing, and what's not relevant. When you build a common rubric to measure writing over the course of months or quarters or years, there's a lot that you're leaving out. There's also a tremendous advantage of naming the things that count. The goal is to keep this teeter-totter balanced.