Fewer, Better Tests Can Boost Student Achievement
Both Democrats and Republicans have submitted proposals to reauthorize the Elementary and Secondary Education Act, the federal law governing K-12 education that has not been revamped since the much-maligned No Child Left Behind Act—the latest iteration of the ESEA—was signed into law in 2002. Among the few things both parties agree on is continuing to require grade-by-grade testing and a new requirement that would focus the draconian consequences that once threatened all schools on the lowest-achieving schools exclusively; in other words, those schools that primarily serve low-income and new immigrant students. This is a big mistake.
Taken together, along with our continued failure to address equity in school resources, these two provisions would virtually guarantee that the overall performance of our students will never equal that of our toughest international competitors and would further widen the gap between the top performers and our disadvantaged students.
Here's why: Americans are addicted to multiple-choice, computer-scored tests, mainly because they are cheap and easy to score. However, these tests drive a rote curriculum that will not produce the skills students need to get and keep good jobs in the 21st century—writing and speaking well, using advanced mathematics, analyzing complex problems, and finding and synthesizing information from many sources for creative problem-solving.
The countries that outperform the United States on international exams spend more than we do to measure and encourage these skills with essay tests and teacher-scored projects. And they can afford to do this because they test much less frequently than we do, typically only two or three times during a student's entire school career.
Although the new Common Core State Standards outline ambitious skills, states worried about the costs of grade-by-grade testing are putting pressure on the state consortia developing new tests—the Smarter Balanced Assessment Consortium and the Partnership for Assessment of Readiness for College and Careers, or PARCC—to lower their costs, even though this would eliminate the kinds of tasks that can measure the most challenging standards. Commercial testing companies are likely to follow suit, with the result that our assessments will change far too little to match those of high-achieving nations.
We get what we measure. This is truer now than ever in the United States because of new federally required accountability systems that tie teachers' employment and some schools' existence to student test performance. By continuing these accountability requirements and the requirements for annual high-stakes testing that effectively force states to use cheap tests, Congress will virtually guarantee that our teachers have strong incentives to teach a curriculum that leaves out the complex skills and knowledge most needed by our students in the modern global economy.
Moreover, by refocusing sanctions on schools serving the most disadvantaged students, Congress is making sure that the students who most need a top-flight education to create a bridge to the jobs of the future will, in fact, receive the most limited curriculum.
If, at this critical moment, the United States chooses to measure 20th-century skills when our top competitors are measuring 21st-century skills, we will effectively be putting a cap on what the American education system will be able to deliver, sealing the fate of our most disadvantaged students and likely of the entire country.
There is a simple fix. Absent major federal investments in much better assessments, Congress should require that states ensure external testing for a single grade at each of three school levels—elementary, middle, and high school—while continuing to report scores separately for vulnerable groups. Without spending any more than they do now, states could employ higher-quality assessments that encourage more productive teaching while reducing the testing burden on students and teachers.
While some will argue that reducing the number of grades tested would reduce accountability and weaken incentives to improve performance, there is no evidence here or around the globe that more testing produces higher achievement.
The record of the top-performing countries is clear—high standards aligned to first-rate curricula and assessments are a necessary step to enable high and equitable achievement. To match the academic performance of these countries, we will have to adopt the other key contributors to their success: fairer systems of school finance, more support for early-childhood education, more coherent and powerful instructional systems, and a whole galaxy of policies aimed at recruiting and training top-notch teachers and fairly distributing those teachers among schools and students. If we are to succeed, we will need to create tests worth teaching to and implement the common-core standards along with common-core supports.
Vol. 33, Issue 07, Pages 24-25