EDUCATION REFORM Beneath the Surface
Part II: Marketizing School Reform

I ended the first article in this series by discussing the Department of Education’s focus on closing old and opening new schools. I’ll begin this second article by discussing the complementary focus on the marketization of school selection.

I’m no expert on markets. Indeed, I mostly rely on Michael Lewis, author of books such as The Big Short and the recently published Flash Boys, to understand them. Come to think of it, it might not be a bad idea for Michael Lewis to devote his next book to exposing the underside of the marketization of school reform. In the meantime, let me try to sort things out a bit.

In New York it all began with the establishment of new small high schools to replace quite deadly large high schools, especially in the Bronx. The plan was simple and quite good—close one bad big school and open up three or four new good small schools. That effort began before the Bloomberg era but it was enthusiastically embraced by the new regime. Early on, it secured the support of many millions of dollars from the Gates Foundation. Over a decade, more than 400 new schools were started. A fair assessment would say that most of the small high schools are much better than the schools they replaced, evidenced by much higher rates of high school graduation. But most of those new schools have not reached the goals they were intended to achieve. Those schools were not simply supposed to be better than really dreadful schools; they were intended to be really good schools. That goal remains elusive.

But, in the meantime, one thing did lead to another. The process of developing and approving new schools morphed into what was called, by analogy with financial investment, a “portfolio” strategy. A portfolio of stocks includes winners and losers; over time, the winners and losers are identified; the losers lose money and the winners make it. When this approach is tried on schools, the kids who attend the loser schools—oops!

Ah! The way you avoid that bad outcome is that you give people choices, so they can identify winners and losers and pick winners. This didn’t actually work out so well with 401Ks. In any case, in a school system with only a few winners and a lot of losers, the choice is often a hollow one. Let’s try to understand the balance of winners and losers. Here in New York City, only about 22 percent of high school graduates are meeting the state’s expectations for being prepared for college—meaning that they would not be required to take remedial courses if they did enroll in college. This is a low bar. But that’s hardly the worst of it. For the just over 350 schools for which data for 2012’s graduates were reported, 10 percent (about 40) of the schools produced half of the college-ready graduates. Many of those schools are exam schools or have selective admissions policies. About half of the high schools produced 15,000 of the 16,000 deemed ready; the other 170 schools produced a total of just over 1,000 (out of over 22,000 students). Depending on how rigorous you want to be, either 90 percent or 50 percent of the high schools are losers.

And then we come to charter schools—perhaps the embodiment of the choice strategy. New York City now has about 180 charter schools, not quite 10 percent of the total number of public schools in the city. They enroll approximately six to seven percent of the city’s public-school children. All but six of the charter schools are operated by non-profit organizations, and state law prohibits any new charter schools being operated by profit-making entities. Until the recent state budget law was passed, charter schools received the same per capita funding as the city’s regular district schools. The city has had to pay $13,527 per charter student, but the new state law increased that amount to $13,777, not a whole lot but symbolically significant. The teachers in most of them are not unionized but there is no bar to union organization—teachers in 22 of them are members of the United Federation of Teachers (U.F.T.). Admission to the charter schools is by lottery and open to any interested family. There are preferences given to siblings of children already in a school and to residents of the local school district. But there are no academic requirements for admission. Charter schools are disproportionately located in school districts with the worst performing regular schools. Two high-performing districts in Queens have no charter schools.

In Harlem, 20 percent of children are now attending more than 25 charter schools. While parents of young children in Harlem have no choice about the zoned school their children will attend, no one is forcing Harlem parents to send their kids to a charter school. In all likelihood, they decided to apply to a charter because they hoped their children would get a better education than they would have in the school they otherwise would have had to attend. Whether charter schools deliver on that promise is a separate matter. But a blanket opposition to charter schools simply does not take seriously the disappointments and fears of the parents of black and Hispanic children, among others.

Because charter schools are freed from many of the institutionalized regularities of the larger school system, it is somewhat easier for them to achieve various kinds of social cohesion—not all of them so enlightened. Perhaps the flagship of charter schools are the KIPP schools—the Knowledge is Power Program, the brainchild of David Levin and Michael Feinberg, two Ivy Leaguers and Teach for America alumni. Their defining mottos are “Be nice, work hard” and “There are no shortcuts.” One critic has described their approach:

Feinberg and Levin seem to have created a steroid and manic version of the traditional classroom. They combine this vision with ongoing psychological interventions intended to breed an unwavering positive outlook among students. The energetic and bureaucracy-busting reformers also borrowed from the inimitable Harriet Ball, whose teaching style offers a culturally-sensitive mashup of gospel, hip-hop and Schoolhouse Rock, which, no doubt, loses some of its effect and charm in the hands of the white T.F.A. teachers that KIPP has recruited since the early days in Houston. It was one of Harriet Ball’s chants … that inspired the name, Knowledge is Power Program:

You gotta read, baby, read.
You gotta read, baby, read.
The more you read, the more you know.
Cause knowledge is power,
Power is money, and
I want it.

Put simply, KIPP’s education philosophy is behavior modification and political indoctrination. KIPP has no coherent academic approach other than a series of gimmicks to inculcate student compliance in drill-like activities. While many charter schools share these kinds of approaches with KIPP, not all do. Some have quite sound educational visions and would never imagine treating kids as KIPP does. Again, some charters appear to be quite successful in comparison to regular district schools but many others appear to be no better and even worse.

There is another objectionable thing that some charters, including KIPP, do: they exclude kids who they come to believe will not be successful—meaning they will not do well on state tests. Apparently, at one point, Geoffrey Canada of the Harlem Children’s Zone simply dismissed an entire class of students from his charter school for this reason.

Which brings us to the next point—the rule of data! At the same moment when then-education chief Joel Klein announced the new era of empowerment described in the first article, he also announced that schools would henceforth be evaluated according to a comprehensive new accountability system that would assess the performance and progress of their students. Thus were born the now well-known Progress Reports, which assign schools grades from A to F. (Informed sources believe that the new leadership at the D.O.E. will retain the progress reports but will eliminate the letter grades).

After all was said and done, the centerpiece of the judgment of schools was the performance of their students on New York State exams—in the case of elementary and intermediate schools, performance on annual tests in English Language Arts and Mathematics; in the case of high schools, performance on Regents exams. The thinking behind the accountability system was fairly simplistic: we want better results; we don’t think that we know how to achieve them; we’ll let individual school principals figure out what to do; we won’t care especially much about what they do or how they achieve their results, and we’ll reward the high performers and punish the poor performers. The data will drive change. The results should not surprise. The writers of the hit TV show The Wire boiled it down: In one episode, a new city teacher, formerly of the Baltimore police, hears how his school will teach test questions; the young man immediately recognizes the dilemma: “Juking the stats ... Making robberies into larcenies. Making rapes disappear. You juke the stats, and majors become colonels. I’ve been here before.”

Most standardized tests are “teachable”—when similar tests are used year after year (and they often are, since that’s the simplest way for the testing companies to promote the “reliability” of the tests). Schools and teachers can craft their instruction around the expected form and content of the test items rather than more broadly around the content and skills that the test is ostensibly intended to sample. When the future employment of teachers and principals, as well as the future existence of a school, is at stake, many cannot resist the short-term advantage of test prep as a dominant instructional strategy and of cheating as a good back-up. The truth of the matter is that test prep models are not nearly as effective in producing high scores, presumably reflecting high levels of student achievement, as would be the use of high-quality curriculum and sophisticated instruction. But far too many teachers and schools don’t necessarily know how to recognize the difference and, if they were freed from the shadow of mandated testing, they might very well use approaches that are not especially more thoughtful than “drill and kill” test prep.

The fundamental emptiness of a strategy grounded in letting principals do whatever they thought necessary to get higher scores was revealed in 2009 when the New York State Education Department recalibrated the scores on its elementary and intermediate school exams and test scores in the city dropped precipitously. A common response from principals to this new specter of unexpected failure was “We’ll have to step up our game!” Not exactly an approach that might lead to more effective teaching!

The specter of data also hangs over teachers as they become subject to more formal evaluations. Most approaches to teacher evaluation rely on standardized test scores for a set percent of a teacher’s overall rating. The balance is determined by student achievement on school-made measures, formal observations of teachers in classrooms and a review of other aspects of teachers’ involvement in the life of the school (such as participation in planning teams). However, the use of test scores has dominated the discussion. In part this is because most states and local districts are making much ado about the use of “value-added” measurements in what they claim is an effort to level the playing field and not to penalize teachers who work with less-successful students. In spite of a lot of razzle-dazzle, the claims by the proponents of “value-added” approaches seem to be flimsy ones—so much so that they have led John Ewing, a prominent mathematician, to accuse its proponents of “Mathematical Intimidation.”*

After all the talk and effort, it increasingly appears that the reforms launched by the Department of Education during the Bloomberg regency have not yielded consistent gains. Higher test scores were often enough found to be based on student performance on predictable tests with cut points set far below what had been expected (if not on outright cheating); high school graduation rates have increased but the majority of graduates are still not deemed ready for college. But, perhaps worse still, as we’ll see in the next part, educational inequality may have been “baked into” the routine operations of the school system even more thoroughly than was the case before the era of reform.

* John Ewing, “Mathematical Intimidation: Driven by the Data,” Notices of the American Mathematical Society (May 2011).


The first article in this series appeared in May’s FIELD NOTES.