I have a complicated relationship with testing. I refuse to pretend that it’s caused no mischief in our schools—narrowing curriculum, encouraging large amounts of ill-conceived test prep, and making school a joyless grind for too many teachers and students alike—but neither can any fair-minded analyst deny that there have been real if modest gains in our present era of test-driven accountability, especially for low-income black and Hispanic children, particularly in the early grades.
What to make, then, of Secretary Duncan’s widely heralded concession that testing is “sucking the oxygen out of the room in a lot of schools” and his offer to states of a year-long delay in making test scores part of their evaluation systems?
“There’s wide recognition that annual assessments—those required by federal law—have done much to shine a light on the places and groups of students most in need of help. Yet in too many places, it’s clear that the yardstick has become the focus,” Duncan wrote in a lengthy blog post Thursday.
We at Fordham have been among those pleading for some reasonable flexibility in this area, particularly as new standards and assessments kick in, so the secretary’s message is welcome. Some states don’t want to shift gears, but others crave a breather while curriculum and pedagogy catch up with newly rigorous expectations. (We’ll save for another day an examination of the constitutional aspects of all this, as Duncan’s department evidently will be offering states waivers from conditional waivers, the statutory basis for which has long been in doubt.)
But what about testing over the long run? Duncan’s soul-searching and candor are laudable and refreshing, but will they do more than defer a larger reckoning on testing’s place in American education? “I want our department to be part of the solution,” the secretary wrote. It’s hard to picture what that solution might be.
The difficulty was made immediately clear by those rushing forward to tell the Secretary what he meant.
“The delays are going to be temporary,” TNTP’s Tim Daly blogged. “Let’s hope states that receive a reprieve don’t take a nap…because there is a ton of work to be done on teacher evaluation systems. Early adopter states have struggled with data integrity, inflated scores, and bias in classroom observations,” he wrote. All of which is certainly true.
NEA President-elect Lily Eskelsen García said the new policy was “common sense,” and, like Daly, noted that it’s just a “temporary measure.” Yet agreement on the word “temporary” is about all the agreement we can expect, perhaps ever.
For Daly and many other reformers insist on no napping when it comes to results-linked accountability for teachers, schools, and students, while García noted (in response to Duncan) that “we must end these absurd practices” of high-stakes decision-making based on student test scores.
The restive mood over testing—and the associated threat to reform—is not limited to teachers. The PDK/Gallup poll released last week shows 54 percent of Americans—a majority now—agree that “standardized tests are not helpful” in letting teachers know what to teach, a figure that jumps to an alarming 68 percent when you count only public school parents.
That’s not, to be clear, a generalized anti-test sentiment, as the same poll shows 80 percent support for college entrance exams; 78 percent for tests to determine promotion from one grade to the next; and 91 percent in support of AP testing to award college credit to high schoolers. Americans, it seems, support testing when it’s in the service of a clear, well-defined outcome. But they don’t seem to regard “standardized” testing a la NCLB and the Common Core in the same way.
Meanwhile, a second poll by Education Next this week shows strong support for other aspects of the reform agenda. Charter schools, tax credits to fund scholarships for low-income students, even vouchers poll well. Fifty-seven percent of Americans support basing teacher salaries in part “on how much their students learn.” The irony is acute: support for the reform agenda is driven by data gleaned from testing, even as testing has become the biggest threat to that agenda.
Secretary Duncan’s reflective take on testing can delay, but cannot resolve, the reckoning that seems to be at hand, and will surely come to a head as Americans get their children’s sobering scores on tests aligned to the higher Common Core standards. Test-based accountability is turning teachers against the Common Core (and presumably against other efforts to raise standards) at the same time as politics is turning the broader public against the Common Core in part by associating it with mindless standardized testing.
Sooner or later, something’s got to give.
It seems fanciful to think we can expect teachers to commit themselves wholeheartedly to reaching higher standards while holding the threat of test-based performance reviews over their heads. It’s equally fanciful to expect a clear, fair, and effective evaluation in the absence of objective, test-driven measures. And please let’s not be naïve: reducing the weight of testing in teacher evaluation will not reduce the dominance tests have come to wield in our classrooms.
What’s it going to be? The most reliable means we have of evaluating performance is deeply unpopular. The popular means are deeply unsatisfying, squishy, and easily manipulated. Our relationship with testing is like holding a wolf by its ears. We can’t hold on and we can’t let go.