For those of you who follow my education-related posts (here is a primer), you know I'm no fan of testing in public schools as it is currently conceived and used. In my view, the cart is firmly before the horse, where the horse of quality education is being pulled by the cart of testing rather than the more appropriate other way around. My concerns are myriad:
- It causes schools to "teach to the test," meaning teachers devote most of their time to preparing students to pass tests rather than educating them;
- Teaching to the test causes a narrowing of the curriculum that neglects other essential subjects beyond those of reading and math that are the focus of testing;
- The current tests don't measure important criteria such as motivation, life skills, critical or abstract thinking, creativity, and decision making;
- Using test scores to evaluate teachers ignores the at-best uncertain results of the "value added" benefits of teachers to student learning (though the idea does have potential);
- Teachers, schools, and states are motivated to game the system (e.g., lower standards, cherry picking data, cheating) to ensure federal funding;
Testing sucks the joy out of both teaching and learning; - Ultimately, test scores have become the end-all, be-all of public education reform rather than just a tool to assess the quality of public education.
But what would happen if schools were to "test to the teach?" In other words, use testing as a means of assessing how well students are learning the curriculum that is being taught to them by their teachers (rather than the curriculum being comprised of what is needed to pass the tests). In fact, as Susan Engel, a noted education researcher, suggests, tests could be developed to measure most everything that students learn, both in terms of subjects (e.g., history, science, and vocabulary) and life skills (e.g., abstract thinking and problem solving). And, the results of these tests could be far more useful tools for improving the quality of education and closing the achievement gap than the current misguided use of testing.










