Incentive programs for schools, teachers and students aimed at raising standardized test scores are largely unproductive in generating increased student achievement, according to a new report researched by an expert panel of the National Research Council.
The report said that standardized tests commonly used in schools to measure student performance — including high school exit exams and tests in various grades mandated by former president Bush’s No Child Left Behind law — “fall short of providing a complete measure of desired educational outcomes in many ways,” according to a summary of the lengthy document.
The report, together with a number of other studies released in the past year, effectively serve as a warning to policymakers in states that are moving to implement laws, with support from the Obama administration, to make teacher and principal evaluation largely dependent on increases in students’ standardized test scores.
The practice doesn’t bring about the kind of student achievement policymakers say is necessary for the United States to compete with the highest-performing countries, according to the 17-member Committee on Incentives and Test-Based Accountability convened by the National Research Council, which is the research arm of the National Academies (including the National Academy of Sciences, the National Council of Engineering and the National Academy of Medicine).
The panelists — who include experts in assessment, education law and the sciences — examined over the past decade 15 incentive programs, which are designed to link rewards or sanctions for schools, students and teachers to students’ test results. The programs studied included high-school exit exams and those that give teachers incentives (such as bonus pay) for improved test scores.
The panel studied the effects of incentives, not by tracking changes in scores on high-stakes tests connected to incentive programs, but by looking at the results of “low-stakes” tests, such as the well-regarded National Assessment of Educational Progress, which aren’t linked to the incentives and are taken by the same cohorts of students.
The researchers concluded that the effects of incentive programs tend to be “small and . . . effectively zero for a number” of such programs.
Gains that were detected were concentrated in elementary grade mathematics and “are small in comparison with the improvement the nation hopes to achieve,” according to the summary.
The researchers concluded not only that incentive programs have not raised student achievement in the United States to the level achieved in the highest-performing countries but also that incentives/sanctions can give a false view of exactly how well students are doing. (The U.S. reform movement doesn’t follow the same principles that have been adopted by the other countries policymakers often cite. You can read an analysis of that by educator Linda Darling-Hammond here.)
Current standardized tests “fall short” of measuring student performance in “important ways,” the summary said.
When incentives lead teachers to concentrate in class on material that will be on a test (a practice known as teaching to the test), understanding of untested material can decline, it said. Constraints on testing — including cost and test length — mean that only a subset of material can actually be tested, it said.
Thus, test results may gave “an inflated picture of learning”of material that a student is supposed to know.
The researchers said that more comprehensive evaluation methods should be developed. But they cautioned that policymakers should not make such an investment at the expense of improvements in other areas of education, including curriculum and instructional methods.
Other studies in the past year have also cast doubt on the effectiveness and reliability of the value-added method of teacher/principal evaluation, which takes student test scores and puts them into a formula that is supposed to factor out other influences and determine the “value” a teacher has brought to a student’s learning.
The method often ignores outside-school factors that can influence how a child does on a test, including lack of sleep, hunger and illness, but even formulas that are said to take these into account are not especially reliable, some experts have said.
So far, state and federal have ignored the evidence. Congress, as it figures out how it wants to rewrite No Child Left Behind — assuming it actually gets around to it — should not.
Follow The Answer Sheet every day by bookmarking http://www.washingtonpost.com/blogs/answer-sheet. And for admissions advice, college news and links to campus papers, please check out our Higher Education page. Bookmark it!