Recently released national scores reveal that high-stakes testing is not helping public education.

Scores from the 2015 National Assessment of Educational Progress, often called the “nation’s report card,” show declines in student test scores in reading and mathematics for the first time since 1990. SAT scores have also gone down.

Students are spending an increasing number of days each year taking standardized tests, dedicating weeks or even months preparing. The Council of Great City Schools just released a report documenting the extent to which students nationwide are being overtested, not only in the number of tests, but also in the time spent on testing and test prep. The report shows no evidence that all of the additional time, attention and resources have led to any significant gains in learning or achievement.

None of this should be surprising. Research was clear even before No Child Left Behind and Race to the Top that more testing would do little to improve education.

Reforms that found solid footing in the civil rights movement, including desegregation and multicultural curriculum, were actually closing the gap between white and black students. But in the past quarter-century, there’s been increasing preoccupation with test scores.

Secretary of Education Arne Duncan will step down in December. His successor is John B. King, whose track record does not suggest a shift in direction. As chancellor of the New York schools, King faced an historic vote of no-confidence by the teachers union and calls for his resignation by parent groups due to his obsession with testing.

The Department of Education continues to call for annual testing and for making high-stakes decisions based on whether or not test scores rise, including whether to fail students, remove teachers, close schools and defund teacher-preparation programs. This is despite testing experts and even the test-makers themselves calling these high-pressure tactics a misuse of test scores.

In just the past few months, more states have shown skepticism over the assessments for the Common Core State Standards and have reversed or delayed earlier plans to require these tests for high-stakes decisions. These are wise decisions because no compelling body of research exists that suggests the Common Core and the tests connected to it will improve achievement or close gaps.

This fall, as Congress works to reauthorize the Elementary and Secondary Education Act, and as the department prepares for a new secretary, the American public should demand a shift in direction. We need to reorient our schools away from testing.

— Kevin Kumashiro is dean of the University of San Francisco School of Education.