Multiple Measures Mess

Title I Monitor has an important story looking at the debate over multiple-measures in No Child Left Behind assessment. This is a longstanding debate in federal policy, since standards got in the federal game, and the kind that plays out behind the scenes among a handful of cognoscenti but has big implications for a lot of folks. Travis Hicks does a good job laying out the issues, and I’m long a fan of more prescriptive language here in federal statutes themselves (a fight that was lost in 2001) but Diane Piche and Cindy Brown lay out the policy reality pretty well. All that said, however, two thoughts: First, smart investment here are key. There is no point in building the best buggy whip factory the world has ever known at the dawn of the automobile era. In other words, assessment is changing fast, and — to mix metaphors — don’t fight the last war. Second, all multiple-measures are not created equal at all and some of the proposals floating around would do nothing to help with the reliability or validity problems and could in fact exacerbate them. And relying mostly on localized assessments make great rhetoric but is a lousy public policy that disadvantages some groups of students.