How can we make standardized testing more valuable? Dump accountability.

Momentum is building to use assessment data to determine what students need to recover from COVID's disruptions.
By: | January 31, 2022
(AdobeStock)(AdobeStock)

Here is a big assessment question for 2022: How will standardized tests bounce back from COVID’s vast disruptions?

Or will they?

Though many students are likely to spend most of this school year learning in person, short-term shifts back to remote instruction and other disruptions could again chip away at the reliability of high-stakes, year-end assessments. Even if the staggering omicron wave is shorter-lived than past surges and schools recover some stability this winter and spring, there will be questions about whether 2022 scores can be compared to those of 2019.

Steady enrollment and, therefore, test participation rates could remain an issue for many districts, says Scott F. Marion, executive director of the Center for Assessment, a New Hampshire-based nonprofit that helps districts design tests that are based on learning goals. “If test scores are a little higher than 2021, which I suspect will be the case, it likely confirms what saw in 2021—that we still have a long way to go,” he says. “And we will have to think about how to analyze the data appropriately to answer questions such as, ‘Is learning accelerating more than typical year-to-year growth?’”

No one expects that standardized tests are going away and their role in the coming years will, in part, be determined by the ongoing push-and-pull between districts leaders clinging to local control and state and federal policymakers who want to exert authority over K-12 via test-driven accountability systems. But momentum is also building to reorient the testing landscape toward using the data to determine what students and schools need to improve, rather than as a tool for punishing districts and educators for deficient performance, says Chad d’Entremont, executive director of the Boston-based Rennie Center for Education Research & Policy. “No one ever says, ‘I hope we don’t know anything about what students are learning,’” d’Entremont says. “The main discussion we should be having is not whether assessments should exist but how can we create systems that are more informative, less biased and more helpful to improving teaching and learning.”

Testing what we already know

The current system of standardized testing doesn’t provide a lot of surprises, such as when results show that students in high-poverty neighborhoods tend to perform lower than do learners in more affluent neighborhoods. “They’re good at documenting large social trends that we already know exist,” he says.


More from DA: COVID learning loss remains unequal despite signs K-12 is bouncing back


On the one hand, some educators and policymakers may be hesitant to press for testing to prevent adding placing more stress and responsibilities on teachers and students still grappling with COVID’s disruptions and trauma. On the other, others in those two groups believe the data generated by testing is needed now more than ever to gauge the pandemic’s impacts.

All education leaders, however, should be working together to set clear goals for their assessment systems. d’Entremont says he hopes that means a shift away from using tests to hold schools and educators accountable.

“If you ask most people, they would say they want information on how students are learning so we can better intervene and customize support so they can reach their full potential,” he says. “Our current assessment system doesn’t do that; it doesn’t provide timely and useful information during the course of the school year.”

Two more crucial changes would be 1.) conducting tests at more regular intervals, focusing on what students have learned most recently and providing data to teachers immediately so they can adjust their lesson plans, and 2.) using more computer-adaptive tests. The latter would tailor questions to students’ demonstrated level of knowledge to provide more exacting and efficient tests, he says. This technology could also help eliminate cultural bias from test questions. Students could actually choose questions that ask about concepts that relate to their life experiences. “We can’t ask them about skiing, for example, if they’ve never been skiing before,” he says.