Bill Hammond: Why New York’s ‘Failing Schools’ Fail — and How We Can Turn the Tide

After all, why mince words about the very worst public schools in the state, where 94 percent of the kids are flunking math and English proficiency tests? If that doesn’t count as failure, what would?

But it’s hard to be quite so judgmental after reading a thought-provoking study by independent researcher John Bacheller, as presented in November at the Rockefeller Institute of Government in Albany.

Crunching numbers from the state Education Department, Bacheller documented a near-lock-step correlation between low test scores and high concentrations of poverty.

For every 10-point increase in the percentage of low-income students attending a given school, that school’s scores in English and math proficiency tests dropped an average of 6.3 points.

Bacheller calculated that poverty levels account for 79% of the variation in test results. This implies that factors such as teaching quality, curriculum design, textbooks, facilities, budgets — hotly debated as they are — account for around 20% of the difference in school performance, combined.

Zeroing in on the 178 schools officially designated as “failing” by the state, Bacheller found high concentrations of poverty in all of them, with 65 percent or more of their students qualifying for free or reduced-price lunch. (That list of struggling schools is now shorter, after state officials recently removed 70 schools that they said had shown progress)

The test scores at these schools were among the worst in the state, with an average of just 6 percent of students passing. But that was only four points less than would have been predicted based on their poverty levels, which Bacheller said was not statistically significant. (The statewide average passing rate has been roughly 30 percent or higher.)

“The recent arguments about the Common Core of knowledge, testing and ‘failing schools’ center about the notion that by improving our schools, we can improve achievement levels,” Bacheller wrote. “But it is also clear that the ability of schools in our central cities to encourage student success is constrained, even swamped, by external factors that they cannot control.”

The connection between high poverty and schools has been highlighted before, including in a 2015 report from the Fiscal Policy Institute. But Bacheller, a self-described “data grubber” with a Ph.D. in political science, brings new statistical rigor to the debate. And having retired after a long career as a fiscal researcher for state government, he has no special-interest ax to grind.

Bacheller said he dug into the data on poverty and education after reading a July article in The New York Times Magazine that described Syracuse as “one of the poorest cities in America” — which is not how he remembered the town where he went to college. Checking census data, he saw a sharpening divide between inner cities with high levels of poverty and the relatively affluent metropolitan areas surrounding them. The same racial and economic segregation was reflected in those cities’ schools, with suburban students scoring relatively well on standardized tests as downtown kids floundered.

Experts and advocates across the spectrum agree with Bacheller’s basic diagnosis – that income and educational outcomes are tightly linked. But Bacheller’s policy prescriptions give people on all sides something to disagree with.

To start, he sees little argument for the multi-billion-dollar increase in education funding being advocated by school officials and teachers unions. “I don’t know that it would be effective,” he said. He noted that Rochester schools have dismal test scores despite spending more than $20,000 a year per student.

But he’s also critical of the state’s policy of labeling schools as “failing” or “struggling” based on low test scores that, in his view, are largely the result of poverty.

“Why pick on some of them and not others?” he asked. “There certainly is variation (in school performance), and poverty doesn’t explain all of it. But it explains an awful lot of it.”

That position runs against the current tide of state policy. Under a law pushed through by Gov. Andrew Cuomo last year, schools in the bottom 5 percent of test scores, or with graduation rates of less than 60 percent, must show improvement within three years or be designated as “failing.” They then face the possibility of a state-ordered management takeover and staff shakeup.

Bacheller argues that boosting anti-poverty programs in those districts would do far more to help their students than any change in school management.

Many education advocates, like Jenny Sedlis of Students First NY, dismiss Bacheller’s conclusions as “baffling.”  

“Without a doubt, poverty and disadvantage impact student performance, but to deny that other factors – such as teacher and school quality – are also significant drivers of student learning flies in the face of decades of empirical evidence,” Sedlis said. “Worse, as poverty is not an easily remedied problem, such a claim condemns disadvantaged students to a lifetime of low achievement.”

Sedlis pointed out that certain charter schools, which are publicly funded but privately managed, have managed to post some of the highest test scores in the state even as they enroll concentrations of low-incomes students.

“To claim poverty as the only significant driver of student achievement – as this report does – is to classify disadvantaged students as unable to learn and improve,” she said. “This is something we refuse to believe.”

In his report, Bacheller acknowledged the high performance of some charters and said their “no excuses” model should be replicated in upstate schools.

Michael Petrilli of the Fordham Institute for Advancing Educational Excellence (and an occasional contributor to The 74) agreed that rating schools based on performance alone is “insane.” But he said the answer to Bacheller’s concern is not to abandon school grading programs, but to make sure they measure how student performance changes over time.

“You can start to distinguish between high-poverty schools that are making a lot of progress for their kids, and therefore should be considered excellent schools, and high-poverty schools where there is very little progress happening,” Petrilli said.

New York’s program does this, to an extent. It starts with a list of the lowest-performing schools — which, predictably, are all high-poverty — but then allows schools to escape designation if they can document student progress.

Another solution to high-poverty schools springs to mind, though it might be the most politically difficult of all: integration.

As Bacheller points out, the factor driving low scores is not just poverty, but concentrated poverty. As a school’s percentage of low-income students goes down, its test scores for all students go up. He found that the wealthier kids’ scores tend to improve more, but that the poor kids do better, too. (The 74 has written more about the benefits of integration in New York City schools)

If only New York could find a way to combine some of its high-poverty inner-city districts with the wealthier, whiter schools that surround them, maybe their students could go from “failing” — or “struggling” — to success.