Eva et al. Flunk the Fairness Test
In the pre-Common Core era, we had a big problem. Most state tests measured minimal competency in reading and math. But we failed to communicate that to parents, so they reasonably thought a passing grade meant their child was pretty much where they needed to be. Little did they know that their kid could earn a mark of “proficiency” and be reading or doing math at the twentieth or thirtieth percentile nationally. Frankly, we lied to the parents of too many children who were well below average and not at all on a trajectory for success in college or a well-paying career.
Playing games with proficiency cut scores provided much of the impetus behind Common Core. States raised standards and started building tests pitched at a much higher level. Most states are giving those tests for the first time right now, though New York and Kentucky made the transition two years ago. As of 2013, New York’s tests were the toughest in the country, according to a new analysis by Paul Peterson and Matthew Ackerman in Education Next, matching—if not exceeding—the performance standards of the National Assessment of Educational Progress.
That may solve the “proficiency illusion” issue. But now we have a new problem. Some education reformers and media outlets are already using the results of the new, tougher tests to brand schools as “failing” if most of their students don’t meet the higher standards. Note, for instance, the Daily News’s special report, “Fight for their Future,” which leads with the provocative headline “New York City is rife with underperforming schools, including nearly two-thirds of students missing state standards.” This line of attack closely resembles the talking points of Eva Moskowitz and Jeremiah Kittridge of Families for Excellent Schools, who both promote the notion that in New York, “800,000 kids can’t read or do math at grade level” and “143,000 kids are trapped in persistently failing schools.”
These statements are out of bounds, and reformers should say so. They validate the concerns some educators voiced all along: that we would use the results of the tougher tests to unfairly label more schools as failures.
Let me be clear: I don’t mind calling schools out as “persistently failing.” Such schools exist, and they should be subject to aggressive interventions, including closure. And I’d be thrilled if they were replaced by high-performing charters like Eva’s Success Academies. But as I’ve argued ad nauseam (and the Shanker Institute’s Matthew Di Carlo has patiently and persuasively explained for years), evaluating schools based on proficiency rates alone is bad math. (Moskowitz and Kittredge define a “persistently failing school” as one in which 10 percent or fewer of the students are proficient in reading and math—or, in the case of high schools, where the same percentage or lower is testing at college-ready levels.)
That’s because passing rates have as much to do with the performance level of students when they enter the school as the amount of learning that happens once they are there. Particularly in a high-standards scenario, it’s quite possible for schools to help students make lots of growth and still not attain “proficiency.” Looking only at annual passing rates, and not where students start, is terrible practice. This is especially true for middle schools and high schools, where students can enter four years or more behind. There’s a non-trivial risk of branding as failures some schools that, despite low scores, are actually doing right by kids.
All Families for Excellent Schools needs to do is look for low-proficiency, high-growth schools. There probably aren’t that many of them. But subtracting these from the “failing schools” equation would go a long way toward demonstrating to educators that we aren’t playing a game of gotcha.
And what about the Daily News’s claim that “nearly two-thirds of students” are “missing state standards”? That’s true, but it’s also to be expected. New York set its cut score to align with its definition of “college and career readiness.” The result of this process was a cut score that defined proficiency at about the seventieth percentile. That might sound high, but it makes sense because we know from NAEP, ACT, SAT, and college remediation rates that only about 30 percent of high school graduates are truly college-ready—defined as being able to arrive on campus and succeed in credit-bearing courses from day one. (We have less information about how many are career-ready, though if we’re talking about careers that require any kind of technical skills or advanced training, it’s probably in the same range.) Ergo, most kids below the seventieth percentile in math or reading are probably not on track for college or a sustainable, well-paying career.
The hope is that, over time, more students will reach the standards as schools raise expectations and improve teaching and learning. (In other words, we’ll shift the bell curve to the right.) And as a result, more students will be college- and career-ready upon graduation from high school. But that’s not going to happen overnight.
The move to higher standards means that we need to recalibrate our rhetoric and, more importantly, our approach to school accountability. In the low-standards days, it was perfectly legitimate to call out schools that couldn’t get all or most of their students to minimal levels of literacy and numeracy. It simply doesn’t work to similarly defame schools that don’t get all of their students “on track for college and career.” It’s a much higher bar and a much longer road.
It’s utopian to think that most American children will master the Common Core standards immediately. It’s defeatist to think that schools can’t do anything to help their students make progress toward that lofty objective. And it’s disingenuous to take either of these extreme positions in this debate.
– Mike Petrilli
This first appeared on Flypaper.