Of all the school reform ideas circulating today, high-stakes student testing has to be about the worst. Unfortunately, instead of slowing down and reassessing the use of these tests, more and more states seem determined to make them the engine of their school-improvement strategies. Apparently they would rather risk a public backlash than be perceived as “backsliding.”
Those who push for high-stakes testing contend that the threat of severe penalties pressures students and teachers to improve performance. Schools are doing poorly, they imply, because students and teachers aren’t working hard enough. This argument brushes aside decades of research that links poor performance to the way schools are organized, operated, governed, and funded. It also ignores the impact that poverty and discrimination have on student performance.
It’s simple-minded to think that high stakes—and the fear they engender—will drive students to learn more. But even if the theory were sound, it’s badly flawed in practice. For example:
- Many states use “off-the-shelf” standardized tests that are not aligned with the standards and curricula they’ve installed in schools. In some cases, high stakes are attached to norm-referenced tests, which rank students according to the performances of their peers, not to any academic benchmark. How logical is that?
- Many middle and high school students—especially those in the most disadvantaged neighborhoods—do not read well and have not had an opportunity to learn the material required to meet the standards set for them. Many will become frustrated and quit school, undoing years of effort to reduce the drop-out rate.
- States are using standardized test scores as the single measure to determine whether a student passes or graduates, which is stupid and unfair. We’ve created a situation where a student can be held back or denied a diploma for failing to answer a single question correctly. Even the test makers acknowledge the limits of their products, cautioning against overreliance on them and advocating the use of multiple measures. Schooling is—or ought to be—about much more than what can be measured by a standardized test.
- Most teachers have not been adequately prepared to teach to new standards. What’s worse, the teachers generally assigned to the most at-risk students are those least prepared for the task.
When introduced in the late 1980s, the blueprint for standards-based reform included a number of crucial building blocks that have largely been ignored. Policymakers leapfrogged over these essentials to stress accountability. Standards advocates originally argued, for example, that states adopting rigorous academic benchmarks must also guarantee that each child receives an education adequate to reach them. But these so-called “opportunity-to-learn standards” were quickly tabled on the grounds that they would be too costly and, therefore, not politically viable.
Standards proponents also insisted that tests used to monitor students’ performances should be aligned with curricula and include more than primitive multiple-choice items. A handful of states have developed such tests, but the majority hasn’t.
What’s more, advocates urged that students be given the time they need to master standards. Instead, high-stakes tests (along with grade-specific standards) actually reduce flexibility and emphasize even more the lock step process of schooling.
The legitimate need for accountability makes it unrealistic to propose eliminating standardized tests. But policymakers and educators should not use test scores to make promotion and graduation decisions until they can guarantee that all students have teachers who are adequately prepared to teach a curriculum that is aligned with state standards. Needy students should also receive additional help and time.
Many organizations and individuals are working hard to fulfill the enormous potential of standards-based reform. It would be a pity to let high-stakes testing derail the movement now.
—Ronald A. Wolk