Editor's note: This testimony was presented at a hearing of the Senate Committee on Health, Education, Labor and Pensions onFixing No Child Left Behind: Testing and Accountability on January, 21, 2015. It additionally appeared in a slightly different form at Education Next.
Chairman Alexander, Senator Murray, Members of the Committee:
Thank you for the opportunity to appear before you today. I would like to begin by congratulating the committee on putting the reauthorization of the Elementary and Secondary Education Act at the top of its legislative agenda for the 114th Congress. Nothing is more important to our nation’s future than ensuring that we provide all children with the opportunity to reach their full academic potential. Congress cannot do that on its own, but it can help by addressing the very real shortcomings of the most recent reauthorization, No Child Left Behind, and restoring the predictability with respect to federal policy that state and local officials need to carry out their work.
As you move forward with this important work, however, I would urge you not to lose sight of the positive aspects of No Child Left Behind. Above all, the law’s requirement that students be tested annually in reading and math in grades 3-8 and once in high school has provided parents, teachers, and other citizens with detailed information about students’ performance in these foundational subjects – and therefore the extent to which they have mastered skills that are prerequisites for other educational goals. This information has called attention to achievement gaps along lines of race, ethnicity and class, across entire states and within specific schools; it has ushered in a new era in education research; and it has made it possible to develop new indicators of schools’ performance based on their contribution to student learning. Research confirms that, by requiring states that had not previously implemented school accountability systems to do so, No Child Left Behind worked to generate modest improvements in student learning, concentrated in math and among the lowest-performing students – precisely those on whom the law was focused.
I say worked in the past tense, however, as the days when No Child Left Behind worked are behind us. As the law’s 2014 deadline for all students to be performing at grade-level approached, its accountability system became unworkable. Far too many schools were identified as under-performing, and the system lost its most critical asset: its credibility.
Recently concerns have also been raised about the amount of time students now spend taking standardized tests. We lack systematic data on the amount of time students nationwide spend taking standardized tests, nor do we know how much would be optimal. A handful of recent district- and state-level audits suggest that students spend 1-3% of the school year taking standardized tests, depending on the grade level, a figure that sounds appropriate given the value of the information they provide and evidence that test-taking itself can support learning. But we also know that some schools test far more than this, and that too many schools devote excessive time to narrow test-preparation activities in an attempt to avoid federally mandated accountability sanctions. The concerns voiced by parents and educators in these schools are legitimate.
But eliminating annual testing requirements is not necessary to address these concerns. Indeed, it would only make them harder to do so.
It is not necessary because federally mandated annual state tests account for less than half of test-taking time—just 32 percent in a recent Ohio study. The rest of test-taking time in Ohio is devoted to state- and district-mandated tests and to new tests developed to implement the teacher evaluation system the state was forced to adopt under the Obama administration’s ESEA waiver program.
It would make matters more difficult because the most important flaw of the No Child Left Behind accountability system is its reliance on the level of student achievement at a single point in time as a measure of school performance. Achievement levels are a poor indicator of school quality, as they are heavily influenced by factors outside of a school’s control. This approach, which is all that is possible under a grade-span testing regime, judges schools based on the students they serve, not on how well they serve them. Performance measures based on the growth in student achievement over time, which are only possible with annual testing, provide a fairer, more accurate picture of schools’ contribution to student learning.
So why did Congress design such a system back in 2002? Well, one key reason was that many states did not yet test students annually, and those that did were often unable to track the performance of individual students over time. That situation has now changed, thanks to No Child Left Behind and related federal investments in state data systems. It would be ironic and, in my view, unfortunate if, in seeking to fix No Child Left Behind, Congress were to recreate the conditions that led to the adoption of an ill-designed accountability system in the first place.
Eliminating annual testing would have other negative consequences:
First, it would all but eliminate school-level information about the learning of student subgroups, as testing only a single grade in each school often results in sample sizes for groups such as English learners or blacks that are too small to yield reliable information for the school as a whole.
Second, it would sharply limit the information available to parents making choices about the school their child attends, whether through open-enrollment or charter school programs.
Third, it would prevent policymakers and researchers from evaluating the effectiveness of new education programs when, as is typically the case, the appropriate research design depends on knowledge of students’ recent achievement.
My main recommendation, therefore, is to maintain the law’s current annual testing requirements, while restoring to states virtually all decisions about the design of their accountability systems, including how schools and teachers are identified as under-performing and what should be done to improve their performance.
The federal government has a critical role to play in ensuring that parents and citizens have good information about their schools’ performance, and good information requires the data that come from annual testing. States should continue to be required to gather this information and to report on it disaggregated by student subgroup.
At the same time, the federal government lacks the capacity to design an accountability system that is appropriate to the needs of each state, and has a poor track record when attempting to dictate the required elements of efforts to improve under-performing schools.
By focusing on improving the transparency of information about school performance and resources, Congress can build upon the successes of No Child Left Behind while learning from its failures.
Thank you again for the opportunity to join you today; I look forward to my fellow panelists’ comments and your questions.
Martin West is an associate professor of education at the Harvard Graduate School of Education, deputy director of Harvard’s Program on Education Policy and Governance at Harvard University, and an executive editor of Education Next.
