As schools continue to deal with the disruptions of the COVID-19 pandemic, the question of testing and accountability looms large. On the one hand, parents and teachers recognize the devastating amount of learning students have lost and the need to identify what their own children do and don’t know. On the other hand, when students have already missed huge swaths of school time and report high levels of depression and anxiety, no one is eager to sacrifice learning time or human connection in order to have kids hunched over tests.
In the longer term, Buckley suggests that educators and policymakers focus their vision on … [+]
getty
That leaves educational leaders and policymakers in a conundrum: They need the information that testing can provide, but without the burdens that testing imposes. What can they do?
Jack Buckley has some timely suggestions. Buckley, who has previously served as Commissioner for the National Center for Education Statistics, senior vice president of research at the College Board, and the senior vice president for research and evaluation at the American Institutes for Research, and who is currently the head of assessment and learning sciences at assessment company Roblox, has issued a new report sketching a vision of how schools can move towards “high-value,” “low-burden” testing. (Full disclosure: The report was published by AEI, where I’m director of Education Policy Studies).
To help tackle the assessment challenge, Buckley sketches a simple but enormously useful two-by-two taxonomy based on whether the information a test produces is high-value or low-value, with value defined as a product of “what information [a test] provides and the timeliness of those data, whether it has additional benefits (such as college credit), and what legal, bureaucratic, or regulatory requirements it satisfies”; and whether the test itself is high-burden or low-burden, generally defined as “a function of how much testing time it requires and how much money it costs.”
According to this taxonomy, the optimal assessment is high-value and low-burden. Problem is, Buckley points out, that type of assessment doesn’t really exist.
While some tests offer high-value or low-burden, they invariably come up short on the other dimension. An example of high-value/high-burden testing that Buckley cites is the Advanced Placement exams, which are rigorous and accepted for college credit—but also pricey and time-consuming. As an example of low-value/low-burden testing, he flags the National Assessment of Educational Progress—the “Nation’s Report Card”—which provides a cost-effective snapshot of how the nation’s schools are doing, but doesn’t yield information useable by teachers or parents.
Unfortunately, Buckley notes, most of the required testing that states administer each year to comply with federal law and evaluate their school falls into the worst category: high-burden and low-value. As Buckley writes, these tests may be “useful for satisfying federal reporting requirements in support of civil rights monitoring and accountability objectives but not much else, at the expense of a lot of time and money that could be better used on instruction or more diagnostic assessment.”
The pandemic’s disruptions allowed state and district leaders to drop much of this low-value, state-required testing. Yet, Buckley observes, high-value/high-burden testing like the AP exams mostly still happened, as families and educators saw their value. As he puts it, “Despite concerns about fairness and an untested design-and-delivery platform, the College Board had an outpouring of public support to . . . offer AP exams after schools shut down in March 2020. The lesson here is that some assessments are so useful that stakeholders will [find ways] . . . to preserve them.”
Anticipating that many educators and families will “likely refuse to return to the status quo” even after COVID-19, Buckley asks what we might do differently. In the near term, he urges policymakers, at a minimum, to seek opportunities to replace low-value/high-burden tests with high-value/high-burden tests. If we’re going to burden students and teachers, Buckley reasons, we can at least do so in ways that are more useful.
For instance, Buckley finds that there’s broad interest in replacing the typical end-of-year state summative assessment with a series of interim assessments given throughout the year. He concedes that “such a redesigned system might indeed still be burdensome in cost and time. But if executed correctly, it could return more timely and actionable data to guide instruction and support improving student achievement.” Plus, Buckley writes, the digital platforms required to employ these interim assessments already exist for the most part, and the federal Every Student Succeeds Act already allows states to substitute interim assessments for summative ones.
In the longer term, Buckley suggests that educators and policymakers focus their vision on developing those elusive high-value/low-burden assessments. He sees an essential role for Washington in incentivizing companies to take on the kind of significant, financially risky research needed to develop these tools.
Much of the past two decades in education has been consumed by debates over whether we need more tests—or fewer. At this moment, when we need more information on how kids are faring and fewer burdens on students and teachers, that debate feels increasingly irrelevant. I’m taken with Buckley’s argument because he has found a way to sidestep that frustrating clash, while offering parents and educators a more practical way forward. Schools could use a lot more of that right about now.
