It is widely known inside education (but much less so outside of education), that we really don’t know what to teach or how to measure learning. Educational researchers will dispute this, as they spend their entire careers defining learning and measuring it. In science that is allowed, and we accept the conclusions of studies, but recognize the limits of those definitions. (It is a troubling trend that some researchers do not understand the limits they place on their studies, and they seek to generalize from their very specific observations to the multifarious and Naturalistic reality of classrooms.)
It was largely in response to this uncertain nature of measuring learning that professional organizations began defining standards and methods of measuring attainment of those standards late in the last century. That trend in education led to the hyperfocus on testing that has characterized education in the 21st century in the United States.
A generation of educators have accepted the argument that the standards represent what students should learn and the tests mandated by government agencies are an accurate measure of the degree to which students have learned what they are supposed to learn. This rationale can be challenged for many reasons, and I have in previous posts on this blog (and I expect I will again); in this post, I focus on yet another reason we should abandon the testing that have come to dominate schooling.
Goodhart’s law is commonly stated as “When a measure becomes a target, it ceases to be a good measure.” In general, there is little doubt in most schools that performance on the mandated tests is the purpose of the school. (This is a reasonable attitude as scores are reported publicly and used for many economic and political purposes; it is true that jobs may depend on scores.) What Goodhart reminds us, however, is that when those scores became the target, they were no longer a good measure of learning.
In a recent incarnation of the standards and testing approach to school, it was argued that the standards represented what students needed to be “career and college ready” and that the tests measured that preparedness. This was an unjustified claim to begin, but when educators’ target became scoring well on the tests, they no longer became a measure of career or college readiness.
Even if the tests scores did predict this readiness, the fact that educators and students scheme methods of artificially improving tests scores, the scores are no longer meaningful. During my years in education, I observed many test-preparation schemes being proposed… far more than I observed effective teaching.
We are, of course, left with a conundrum. In the social landscape in which we are going to politicize measurements of learning, they will become (immediately) the target, thus they will cease to be effective measures of what we want school to accomplish.