Thompson: An Even Sadder Tale of D.C. Common Core Testing
The Washington Post’s Emma Brown, in D.C. Mulls Common Core Test Switch, explains that four years ago the D.C. schools opted for the PARCC Common Core Test rather than the Smarter Balanced assessment. Back then, little was known about the ways that the assessments would differ. Now, a powerful case can be made that the district should switch to the Smarter Balanced test.
If Common Core tests are necessary, I'd say, in an urban district the case for Smarter Balanced is overwhelming. Arguments against the transition to the more appropriate tests are worrisome.
Brown links to the blogger Ken Archer at Greater Greater Education, who has access to the minutes of a meeting of the Office of the State Superintendent of Education (OSSE). The OSSE administers the district’s tests and it is open to a change away from PARCC. Archer reports that the "OSSE discussed their intentions to engage in a series of stakeholder discussions with regards to the choice of common core next generation assessments.”
But, Chancellor Kaya Henderson has a disturbing reason for opposing the seemingly better test. Henderson opposes a transition because “teachers unions would see it as an opening to attack the Common Core and testing in general.”
The best reason for switching to the Smarter Balanced test is that it is a computer-adaptive assessment. Adaptive testing is one of the promising technologies that were undermined by No Child Left Behind. Adaptive assessments adjust the questions asked based on the test-takers’ ability to handle tougher or easier questions. They could be essential in helping 8th graders with 4th grade skills so they don't give up and drop out of school when standards are abruptly raised.
Adaptive tests provide a better measurement of students’ progress. They reduce the incentive for gaming the system by focusing on “bubble kids” or students who are near-proficient, and ignoring the lowest performing students. Moreover, the Smarter Balanced assessments require less testing time.
With Common Core, the advantage of adaptive tests should be obvious. Repeated failure does no good for anyone. The rate at which standards can be raised is determined by the success of teachers in rebuilding low-performing students’ confidence after they face inevitable setbacks. If adults do not demonstrate exquisite patience, kids are likely to dropout in mass.
Brown and Archer explain that the PARCC test, however, is a "fixed-form" assessment. It can be reliable for assessing students who are on grade level, but it is not very good at measuring students far above or below grade level.
So, if the purpose of testing is learning, not just punishment, why would a testing consortium reject adaptive tests when imposing college readiness tests on a low-performing urban district? The question is doubly important in D.C. where gentrification is increasing the already huge gaps between top and bottom performers.
The director of policy, research and design at PARCC said that states in its consortium “made a deliberate decision not to use an adaptive test because of concerns about equity: They were afraid an adaptive test would cause teachers to lower expectations for struggling students because they knew those students wouldn’t have to answer hard questions on the end-of-year test.”
So, once again, teacher-blaming and "gotcha" is driving policy.
I must make one thing clear. I once supported Common Core standards, but it never occurred to me that its advocates would continue to pursue high-stakes tests “worth teaching to.” In theory, I knew, stakes could have been attached to adaptive tests but such an effort would have difficulty passing the smell test. Unless everyone is treated the same, it would be difficult to sustain sanctions for students or adults using tests results, and even adaptive tests become less accurate with students who are “approximately one year above or below grade level.” Moreover, in an era of so-called accountability, it would be just as attractive to game adaptive testing (offering two tries at true and false questions, for instance) as it has been to abuse the once-promising "credit recovery" system and other assessments. But, adaptive assessments are far better tests to teach with.
I once assumed that Common Core advocates understood the need to choose between higher standards and better teaching tools, and the ratcheting up of primitive test-driven accountability. This sorry tale of D.C.’s commitment to the punitive is just one more example of why I finally realized Common Core is doomed.-JT(@drjohnthompson) Image via.