The Future of Testing

Critical Care

The era of high-stakes testing is over, but what comes next?

Del Stover

Say what you will about No Child Left Behind (NCLB), this landmark legislation left its mark—although not always for the best. For all its good intentions, NCLB will be remembered for ushering in an unprecedented era of high-stakes testing—and for saddling public education with a federal mandate on school accountability that had all the subtlety of a sledgehammer.

Yet, the worst is over. Better times are ahead. Recent years have seen state and federal policymakers steadily retreat from NCLB’s flawed accountability model. Test scores are no longer the sole gauge by which schools are judged. Nor are schools held hostage to arbitrary adequate yearly progress (AYP) goals.

In their place, the nation has moved to implement a more nuanced and fairer approach to school accountability. In Georgia, for example, the College and Career Ready Performance Index (CCRPI) relies on an assortment of data to evaluate school performance. It looks at the academic growth of students over the course of a year, whether schools are paying attention to chronically absent students, and, at the high schools, the focus is on graduation rates and access to advanced courses.

That’s not to say Georgia’s accountability system is perfect. Although the sheer scope of data is useful in analyzing a school’s strengths and weaknesses, the state uses a complicated formula to convert the data into a composite numeric score for each school that’s intended to help the public. But, says George Kornegay, superintendent of the Thomas County Schools, it instead offers parents and community members an overly simplistic picture of their schools’ quality.

Still, the performance index is a vast improvement over how schools were judged a decade ago, and that’s good news for school leaders across the nation.

“We’ve observed quite a large improvement in state accountability over the last four to six years,” says Scott Norton, strategic initiative director for standards, assessment, and accountability for the Council of Chief State School Officers (CCSSO). “I think you’ll see the impact at the local level.”

Nothing illustrates this progress more than passage of the Every Student Succeeds Act (ESSA). Although it maintains a federal mandate to measure and improve student academic achievement, ESSA codifies the nation’s retreat from NCLB by mandating some of the very policy changes put in place by Georgia and other states. For example, the law requires state accountability systems to evaluate schools using a variety of criteria.

In a major policy victory for school boards, ESSA also includes language, originally proposed by NSBA, to restore to states and school boards more responsibility over education decisions. The law requires states to engage stakeholders, such as school boards, in the development of state accountability plans and strengthens local accountability by giving school districts more control over implementing interventions for low-performing schools. No longer are many of these decisions dictated by fiat from the nation’s capital.

“We cannot underestimate the sea change in policy that ESSA represents,” says NSBA Executive Director and CEO Thomas J. Gentzel. “The law redefines the federal role in education policy and reaffirms the importance of local governance. This is going to give school boards important new flexibility in how they meet their responsibilities to hold schools accountable for student learning.”

No one is going to benefit more from these changes than school boards serving high-poverty communities. For much of the past decade and a half, many schools were unfairly labeled as “failing” because of low tests scores—and suffered the brunt of well-meaning but blunt interventions.

The problem with these early policy directives is that both research and experience show that students living in poverty face significant obstacles to learning—and their test scores reflect this reality. Yet, for years, very little allowance was made in state and federal policy to acknowledge what clearly was beyond the control of local schools.

It was a complaint raised even as NCLB was being drafted by Congress, but once signed into law, it took years for federal policymakers to acknowledge the extent of their mistake—and the harm of NCLB’s unrealistic expectations for improvement. Only in 2006 did the U.S. Department of Education finally begin to grant waivers allowing states to use academic growth as part of determining AYP goals.

Although not one of the first states to receive a waiver, Colorado quickly joined the bandwagon. Today it not only looks at how well schools advance student academic growth over a year, but also tracks their progress in improving the proficiency of English language learners. “The Colorado growth model became a kind of template for many states about how to do growth,” says Jeremy Meyer of the Colorado Department of Education.

A growth model is an important evolution in accountability. For all of its flaws, NCLB was a step forward in that it forced schools to disaggregate student data by race, income, and other categories. It also held schools accountable for how they served these historically underserved subgroups of students.

But many argue that test scores—even with academic growth weighed into the mix—still offer a limited picture of how well schools educate children. “Test scores are a measure of student performance, not school performance,” says Matthew Di Carlo, a senior research fellow at the Albert Shanker Institute in Washington, D.C. “They are the crudest way to judge schools.”

That realization has led policymakers to explore other measures of school success. In Indiana, the state’s annual performance report includes data on teacher certification. Connecticut takes into consideration student access to the arts and physical fitness. California’s assessment system gives weight to the resources of a school, student and parent engagement, school climate, and access to “a broad course of study.”

Not all of these nonacademic factors figure into ratings or scores posted by states. But most agree that a more nuanced look at school performance is needed. Some have voiced concerns that downgrading academic data raises the risk that areas of poor performance may go overlooked. It is a possibility that cannot be dismissed entirely. However, it’s worth noting that ESSA still requires states to put a strong emphasis on test data—and that means low-performing schools won’t avoid scrutiny.

What will make or break today’s accountability systems is whether policymakers dig down into the data or make the mistake of NCLB by drawing quick conclusions about interventions or sanctions, says Ben DeGrow, director of education policy for Michigan’s Mackinac Center for Public Policy.

“It’s not enough to just look at test scores and say, ‘Oh, these schools are bad.’ We have a few high-poverty schools that are beating the odds. If you’re making decisions about [intervening in] schools, you don’t want to target the ones that are meeting or beating expectations.”

Nor do you want to make excuses for schools where students are academically underperforming, Di Carlo adds. Although the impact of poverty certainly should be acknowledged, some of the challenges of high-poverty schools—less-qualified teachers, staff turnover, fewer resources, or limited access to advanced coursework—are the consequence of policy decisions for which local and state officials should be held accountable.

Ultimately, Di Carlo says, “You have to make sure you’re actually looking at school performance. You don’t want to judge schools based on the students they serve, but how well the schools serve them.”

To judge schools, state policymakers need robust assessment tools. However, development of the latest generation of standardized tests has faced its challenges over the years, much as have school accountability policies.

It was only six years ago, for example, that 46 states agreed to Common Core State Standards and established two testing consortia—the Smarter Balanced Assessment Consortium and the Partnership for Assessment of Readiness for College and Careers (PARCC)— to develop a set of new standardized tests that promised to be more rigorous and comprehensive than existing state tests.

It was an ambitious initiative that, in subsequent years, fell victim to a political backlash against Common Core and led more than two dozen states to back away from the standards-based tests. This year, only 20 states and the District of Columbia are expected to administer the Smarter Balanced or PARCC tests, according to Education Week.

In other states, students will be taking tests developed by state education departments or purchased from private testing groups. Some states are cobbling together tests from a hybrid of sources. Originally committed to the PARCC tests, Massachusetts gave school districts the option in recent years to use either the PARCC or the well-regarded Massachusetts Comprehensive Assessment System (MCAS) test. This year, the state will use a hybrid test, drawn from both assessment systems, and sometimes referred to as the MCAS 2.0.

One assessment model being piloted in New Hampshire allows teachers to use student classroom work to help determine students’ levels of competency. The goal of the initiative, the Performance Assessment of Competency Education (PACE), is to eliminate excessive testing by using local grading for accountability purposes. To ensure the validity and reliability of this approach, however, state assessments still are given once in elementary, middle, and high school, and comparability studies are conducted.

Another strategy that’s finding a niche nationwide is the use of the ACT and SAT. At least a dozen states mandate or allow the use of one of these tests to satisfy some federal accountability requirements. Some are spurred by a desire to test the college readiness of students. Others seek to free themselves from the hassles of developing or administering another test.

Using the ACT or SAT also allows states to ease simmering opposition among students and parents to the amount of testing they’re experiencing in their schools, says Kristen Amundson, president of the National Association of State Boards of Education.

In recent years, students have staged walkouts to protest the high stakes and time demands of testing. In 2015, more than 200,000 students in New York state elementary schools “opted out” of state exams. Such political pushback has prompted several states to begin cutting the number of standardized tests mandated for accountability.

Although more college admissions offices are dropping the tests as a requirement of student applications, Amundson says the states’ use of the ACT or SAT is proving a useful addition because it “eliminates the entire opt-out conversation at the high school level because parents know their kids will want to take either the ACT or SAT, so why not let the state pay for it?”

So much has changed over the years in both the assessments and accountability formulas of states. For the most part, school leaders acknowledge a generally positive evolution. But this change has come at a price. It would be nice, some say, if policymakers could ease up on the constant tweaks and changes.

“Improvement is good, and you have to improve your tests, say, every decade or so,” says Rob Stein, superintendent of Colorado’s Roaring Fork School District. “But one of our challenges is that we’re switching horses mid-stream. We just switched to PARCC a few years ago. We need consistent year-to-year data for us to see if our efforts are working.”

Some states are seeing yearly changes to their accountability formulas, with the result that schools can see their state rating climb or fall for reasons that have nothing to do with student performance. That creates a public relations challenge for school districts, Kornegay says.

“The CCRPI uses a very complex formula, and every year they change how that formula works,” he says. “So, you can’t compare one score to the previous year. But when you try to explain to the public that your schools went down in part because the state is counting overall achievement less and the achievement gap closure more, the average citizen will say, ‘That’s just sour grapes.’”

That’s a legitimate complaint, but it’s unrealistic to expect anything but change for a while, as many states are still working to submit their plans to implement ESSA’s accountability mandates. What’s more, say state officials, the complexity of today’s assessment and accountability measures—and the state’s responsibility to respond to identified problems—makes it almost inevitable that change is going to continue.

Allison Timberlake, Georgia’s director of accountability, is sympathetic to these concerns, but her state already is trying to respond to local concerns that the state’s numeric score for schools doesn’t provide the public with a nuanced look at school performance. It may be possible, for instance, to present the complex data used in developing a school’s score in a more user-friendly format.

And that would mean more changes, she says. “We are trying to make our tweaks and revisions to do a better job of reporting. Our intent is a more comprehensive accountability system.”

One thing seems clear: Although state policies regarding standardized testing and accountability aren’t perfect, they are a lot more sensible and fair than at any time in recent history. What’s more, local school leaders, empowered by ESSA, will have more say about the school improvement efforts targeting low-performing schools.

Yet it’s anyone’s guess what impact all of this will have. People already are raising concerns that some state officials are too eager to close low-performing schools, expand charters, or intrude into local decision-making—and that schools serving high-poverty communities undoubtedly will feel accountability measures target them for scrutiny.

School leaders cannot expect that all this policy work is going to work wonders, says Kevin Welner, director of the National Education Policy Center at the University of Colorado-Boulder. History doesn’t show that accountability turns around schools; at some point, someone must go into the schools, talk to the teachers, talk to the students, and observe classes to see what’s needed.

“Is ESSA going to drive us to realistic, improved accountability and assessments [and] help us not only figure out how well students are learning, but also how to drive school improvement?” he asks. “It’s probably going to move things in that direction ... but I can’t say it will, by itself, get the job done.”

Del Stover
senior editor of American School Board Journal


Go to top