New York got some tough education news last week: Proficiency on standardized tests, given to students in grades three through eight, is down. Way down. Statewide, the pass rate for math fell from 86 percent to 61 percent; in reading, it plummeted from 77 percent to 53 percent. In New York City, the rate in math dropped from 82 percent to 54 percent; in reading, it fell from 69 percent to 42 percent.
The state recalibrated its tests this year, making them tougher and raising the scores needed to pass. (Last year, for instance, a fourth-grader needed 37 out of 70 points to be deemed proficient; this year, the number rose to 51.) So a drop was expected—thought perhaps not one of such magnitude. For the past few years, however, top education officials—particularly New York City Mayor Michael Bloomberg and schools Chancellor Joel Klein—have pointed to high proficiency rates as proof that their education reforms were working. Now, critics of standardized testing are claiming vindication, saying last week's news proves that tests are too unreliable to use when assessing what students are learning, and how well teachers are teaching. “Will anyone be accountable for the hundreds of thousands of children’s educational plans that were devised with nothing but phony data to use?” the blog Schools Matter asked. In response, Bloomberg and his supporters are defending New York City’s Department of Education by pointing out that, while proficiency has plummeted, students’ raw scale scores have not. They also note that the city has made gains on national tests. (Some critics, however, dispute these claims.)
Who’s right? It’s clear that New York clearly has a lot of work to do, since plenty of students aren’t doing as well as everyone once thought. And Bloomberg should acknowledge that his previous boasting about student proficiency rested on bad, inflated numbers. But this latest news also isn’t the indictment of standardized testing that critics are making it out to be.
On the contrary, New York is a case study in the value of testing, once it’s done right. As many education experts have argued for years, most states’ standardized tests are not good indicators of student achievement. Because No Child Left Behind (NCLB) allowed everyone to create their own tests and set their own proficiency markers, but threatened to punish schools that didn’t measure up, many states watered down their tests’ difficulty and lowered the thresholds students needed to clear. The product, as Secretary of Education Arne Duncan has put it many times, has been a “race to the bottom,” where states strive to make their tests easy and their scores high. By recalibrating its tests, New York produced a more accurate picture of where the state’s students stand. And, now that the tests are more meaningful, they can be used to assess and redirect what’s happening in New York’s classrooms.
Better still, the movement to improve testing goes beyond New York. Well more than half of the states, New York among them, have agreed to adopt common standards, created through a joint project between the National Governors Association and the Council of Chief State School Officers. A study released this month by the Fordham Institute found that the common English standards are better than those that 37 states now use, while the math standards are better than those that 39 use. Once the common standards are officially implemented in classrooms, states will have to create new, hopefully better tests to match them.
If the states follow through with this—and, given their recession-limited resources, it’s not clear they will be able to—they’re likely to see their proficiency rates drop, just like New York’s did. When that happens, people should remember that this shock to the system is a good thing, since a better assessment of student achievement is an important, early step in broader education reform. “It’s not fair to freak out about the results,” says Kevin Carey, policy director of Education Sector, a D.C.-based think tank. “We—the public, policymakers, the media—all need to develop more sophistication about interpreting test score results, which are with us to stay.”