In 2009, the city’s accountability system produced bizarre results. An amazing 84 percent of 1,058 elementary and middle schools received an A (compared with 23 percent in 2007), and an additional 13 percent got a B. Only twenty-seven schools received a grade of C, D, or F. Even four schools the state said were “persistently dangerous” received an A. The Department of Education hailed these results as evidence of academic progress, but the usually supportive local press was incredulous. The New York Post called the results “ridiculous” and said, “As it stands now, the grades convey nearly no useful information whatsoever.” The New York Daily News described the reports as a “stupid card trick” and “a big flub” that rendered the annual school reports “nearly meaningless to thousands of parents who look to the summaries for guidance as to which schools serve kids best.”33
The debacle of the grading system had two sources: First, it relied on year-to-year changes in scores, which are subject to random error and are thus unreliable. Second, the scores were hugely inflated by the state’s secret decision to lower the points needed to advance on state tests. Consequently, the city’s flawed grading system produced results that few found credible, while the Department of Education was obliged to pay teachers nearly $30 million in bonuses—based on dumbed-down state tests—as part of its “merit pay” plan.34
How could parents make sense of the conflicting reports from the city, state, and federal accountability systems? Should they send their children to a school that got an A from the city, even though the state said the same school was low-performing and persistently dangerous? Should they pull their child out of a highly regarded neighborhood school where 90 percent of the kids passed the state exams but the city gave it an F? The city had no plan to improve low-performing schools, other than to warn them that they were in danger of being closed down. Shame and humiliation were considered adequate remedies to spur improvement. Pedro Noguera of New York University observed that the Department of Education failed to provide the large schools with the support and guidance they needed to improve. “They don’t have a school-change strategy,” Noguera said. “They have a school-shutdown strategy.” Chancellor Klein acknowledged that opening and closing schools was an essential element in the market-based system of school choice that he preferred. He said, “It’s basically a supply-and-demand pattern. . . . This is about improving the system, not necessarily about improving every single school.” But there was no reason to believe that closing a school and opening a new one would necessarily produce superior results; in fact, half of the city’s ten worst-performing schools on the state math tests in 2009 were new schools that had been opened to replace failing schools.35
Having promised to make dramatic improvements in the school system, both Mayor Bloomberg and Chancellor Klein pointed with pride to the gains recorded on state tests, calling them evidence of historic change. When the mayor’s reforms were launched in September 2003, 52.5 percent of fourth-grade students were at levels 3 and 4 on the state tests in reading. By 2007, that proportion had risen to 56.0 percent, a gain of 3.5 percentage points (this was a much smaller increase than in the four years from 1999 to 2003, when the scores rose by 19.7 percentage points). In eighth-grade reading, where there had been no progress in the previous years, the proportion of students meeting state standards grew from 32.6 percent to 41.8 percent, an impressive improvement. In mathematics, the Bloomberg administration celebrated an increase in the fourth-grade proficiency rate from 66.7 percent in 2003 to 74.1 percent in 2007, a gain of 7.4 percentage points (again, this was a smaller gain than in the four years before mayoral control, when the number climbed from 49.6 percent to 66.7 percent, or 17.1 percentage points). Eighth-grade scores in mathematics shot upward from 34.4 percent meeting standards in 2003 to 45.6 percent in 2007.36 Laudatory articles celebrating the Bloomberg miracle appeared in Forbes, The Economist, Time, Newsweek, U.S. News & World Report, and USA Today. The stories reported the remarkable transformation of the New York City public schools. And the city’s schools won the Broad Prize in 2007, because of the improved test scores.
It was a shock, therefore, when the federal NAEP released reading and math scores for eleven cities, including New York City, in November 2007. The NAEP scores for 2003-2007 encompassed the first four years of the new regime and provided an independent check on the city’s claims of historic gains.
On the NAEP, students in New York City made no significant gains in reading or mathematics between 2003 and 2007, except in fourth-grade mathematics. In fourth-grade reading, in eighth-grade reading, and in eighth-grade mathematics, the NAEP scores showed no significant change. Nor was there any narrowing of the achievement gap among students from different racial groups. Except in fourth-grade mathematics, there were no gains for black students, white students, Asian students, Hispanic students, or lower-income students. The New York Times published a front-page story about the NAEP report, titled “Little Progress for City Schools on National Test.”37
The New York City Department of Education responded to the stagnant results on the federal tests with a press release claiming that “New York City students made impressive gains” on NAEP. The flat scores on the national tests, the chancellor explained to reporters, reflected that students prepared for the state tests, which were aligned with the state standards. But both the state and the federal tests assessed generic skills in reading and mathematics, not specific content (such as works of literature); the skills should have been transferable. If they were not transferable and were useful only for taking state tests, then students were not prepared to read college textbooks, job-training manuals, or anything else that was not specific to the state tests.38
Using private funding, the city launched a publicity blitz to proclaim its increased test scores and graduation rates. The graduation rates were even more malleable than the test scores, as there were so many ways to adjust them up or down. It all depended on which students were counted or not counted. Does one count only students who graduate in four years or students who take more than four years? Does one count GED diplomas that students obtain outside of regular school? What about August graduates?
According to the state, the New York City graduation rate increased from 44 percent to 56 percent between 2003 and 2008. On its face, this was an impressive improvement, but the rate was inflated in various ways, such as excluding students who left city high schools without a diploma and were counted as “discharges” rather than dropouts. (Many discharges would be considered dropouts by federal standards.) The graduation rate was also artificially increased by a dubious practice called “credit recovery,” a covert form of social promotion for high school students. Under credit recovery, students who failed a course or never even showed up for it could get credit by turning in an independent project, whose preparation was unmonitored, or by attending a few extra sessions. A principal told the New York Times that credit recovery was the “dirty little secret of high schools. There’s very little oversight and there are very few standards.”39
Among those who did graduate, many were poorly educated. Three-quarters of the city’s high school graduates who enrolled in the two-year community colleges of the City University of New York were required to take a remedial course in reading, writing, or mathematics; this was an improvement over the statistics for 2002, before mayoral control, when 82 percent required remediation. But in the earlier period, no one claimed that social promotion had ended.40
Testing was always important in the New York City schools, but it assumed even more importance in the age of data-driven accountability. The DOE centralized admissions to gifted and talented programs, presumably in the interest of equity, requiring all applicants to take the same standardized intelligence tests; only those who reached the 90th percentile gained admission to a gifted program. But the new approach halved the number of children in such programs and halved the proportion of African American and Hispanic children accepted from 46 percent to only 22 percent .41 Any education re
searcher could have predicted this result, because children from advantaged homes are far likelier to know the vocabulary on a standardized test than children who lack the same advantages.
Because of its concentration on raising test scores in reading and mathematics, the Department of Education consistently paid less attention to other subjects. The media, too, closely followed reading and math test scores but ignored such subjects as science and social studies, even though the state tested these subjects. When science and social studies were tested in 2008, twenty-eight of New York City’s thirty-two districts placed in the bottom 10th percentile of the state’s districts in science, and twenty-six districts were in the bottom 10th percentile in social studies. In fact, eighteen districts scored in the bottom 5th percentile statewide in both science and social studies. Not a single district scored at or above the 50th percentile in science, and only one (District 26 in Queens) exceeded the 50th percentile in social studies.42 This was a sobering reflection on the narrowness of what was taught in New York City’s public schools. But no one noticed or cared, because those subjects were not part of the city, state, or federal accountability programs. Thus, as the city concentrated intently on raising test scores in reading and mathematics, the other essential ingredients of a good education were missing.
Test score gains do not always last. In 2005, when test scores rose sharply across the city on state tests (as they did in other urban districts in the state), Mayor Bloomberg “trumpeted the results as an election-year affirmation of his stewardship of the public schools,” as the New York Times described it. The mayor and Chancellor Klein made “a triumphant visit” to P.S. 33 in the Bronx to praise the school’s astonishing gains. The proportion of fourth graders meeting state standards on the reading test more than doubled, rising by a staggering 46.7 points in a single year to 83.4 percent. The mayor lauded the good work of the principal. Soon after the principal’s star turn, she retired with a $15,000 bonus, which added $12,000 a year for life to her pension. The following year, P.S. 33’s astonishing gains evaporated. The meteoric rise and fall of test scores at the school was mysterious.43
Psychometricians are generally suspicious of dramatic changes in test scores. New York University’s Robert Tobias, who was testing director of the New York City schools for thirteen years, was skeptical about sharp gains in test scores and their relation to test-preparation activities. At a City Hall hearing in 2005, Tobias criticized “unhealthy over-reliance on testing as a facile tool for educational reform and political advantage” and said, “Much of this test preparation is not designed to increase student learning but rather to try to beat or game the test.”44
In the future, it is certainly possible that test scores and graduation rates will continue to rise. Or maybe they won’t. Of course, it is better to see scores go up than to see them fall. But when the scores are produced by threats of punishment and promises of money, and when students cannot perform equally well on comparable tests for which they have not been trained, then the scores lose their meaning. Scores matter, but they are an indicator, not the definition of a good education.
Mayoral control in New York City had a mixed record. State test scores went up, and spending went up even faster. From 2002 to 2009, the overall budget for public education grew from $12.7 billion annually to $21.8 billion.45 With such a huge jump, New York City’s successes may have been a testament to the value of increased school spending. Did mayoral control bring greater accountability? No, because there was no way to hold the mayor or the chancellor accountable. Standing for reelection once every four years is not a sufficient form of accountability for the mayor, especially when there are so many other issues for voters to consider. The chancellor answered only to the mayor, so he could not be held accountable either. When there were major foul-ups, such as the misrouting of school buses by Alvarez & Marsal in the dead of winter, no one was held accountable.
Mayoral control is not a guaranteed path to school improvement. On the 2007 NAEP, the cities with the highest scores were Charlotte, North Carolina, and Austin, Texas, neither of which had mayoral control. And two of the three lowest-performing cities—Chicago and Cleveland—had had mayoral control for more than a decade. Clearly many factors affect educational performance other than the governance structure.
No governance reform alone will solve all the problems of the schools. A poorly constructed governance system, as New York City had during the era of decentralization from 1969 to 2002, can interfere with the provision of education. But absolute control by the mayor is not the answer, either. It solves no problems to exclude parents and the public from important decisions about education policy or to disregard the educators who work with students daily. Public education is a vital institution in our democratic society, and its governance must be democratic, open to public discussion and public participation.
CHAPTER SIX
NCLB: Measure and Punish
THREE DAYS AFTER HIS INAUGURATION in 2001, President George W. Bush convened some five hundred educators in the East Room of the White House to reveal his plan to reform American education. The plan, which he called No Child Left Behind, promised a new era of high standards, testing, and accountability in which not a single child would be overlooked. It was ironic that the Bush plan borrowed its name from Marian Wright Edelman of the Children’s Defense Fund, who intended it to refer to children’s health and welfare, not to testing and accountability. Edelman’s fund trademarked the slogan “Leave No Child Behind” in 1990 as a rallying cry for its campaign to reduce the number of children living in poverty.1
The White House meeting was a thrilling event, as are all events at this grand and glorious mansion. Visiting the White House always gives me goose bumps. I went to the White House for the first time in 1965, when Lyndon B. Johnson was president, for a formal state dinner honoring the president of a small Caribbean nation (I was there because my then-husband was active in Democratic politics). Dressed in a long evening gown, I was escorted inside by a stiff-backed Marine in full dress uniform. Nearly a decade later, I was invited to discuss education issues at a small luncheon with President Gerald Ford, along with sociologists Nathan Glazer and James Coleman. In 1984, I was one of about forty educators invited to meet with President Ronald Reagan in the Cabinet Room. A couple of times, when I was an assistant secretary of education, I met President George H. W. Bush (at our first meeting, in an irreverent mood, I pulled up a chair next to him behind his desk in the Oval Office so I could get a great picture, and he cheerfully obliged me). During the Clinton years, I was invited to the White House on several occasions to discuss Bill Clinton’s education initiatives. I once was invited to watch him address hundreds of high school students in Maryland, and he had them cheering as he urged them to do their homework, study harder, and take tougher courses. It was an amazing demonstration of personal charisma.
So, on January 23, 2001, when the new President Bush presented his plans for school reform, I was excited and optimistic. The president pledged that his focus “would be on making sure every child is educated” and that “no child will be left behind—not one single child.” No doubt everyone in the room agreed with that sentiment, though no one was quite certain how it would happen. The president described his principles: first, that every child should be tested every year in grades three through eight, using state tests, not a national test; second, that decisions about how to reform schools would be made by the states, not by Washington; third, that low-performing schools would get help to improve; and fourth, that students stuck in persistently dangerous or failing schools would be able to transfer to other schools.
These four principles, described in a concise 28-page document, eventually became the No Child Left Behind legislation, a document of nearly 1,100 pages. NCLB, as it came to be known, was the latest iteration of the basic federal aid legislation, known originally as the Elementary and Secondary Education Act of 1965.
Large bipartisan majorities in Congress approved NC
LB in the fall of 2001. Under ordinary circumstances, Republicans would have opposed the bill’s broad expansion of federal power over local schools, and Democrats would have opposed its heavy emphasis on testing. But after the terrorist attacks of September 11, 2001, Congress wanted to demonstrate unity, and the education legislation sailed through. The Senate passed the law by a vote of 87-10, and the House of Representatives approved it 381-41. Senate Democrats voted 43-6 in its favor, and House Democrats endorsed it 198-6. Senate Republicans voted affirmatively 44-3, as did House Republicans by 183-33.
The legislative leaders of both parties stood proudly with the president as he signed NCLB into law on January 8, 2002. Democrats liked the expansion of the federal role in education, and Republicans liked the law’s support for accountability and choice (although the law did not permit students to take their federal funding with them to private schools, as many Republicans wanted). Republican John Boehner of Ohio called the law his “proudest achievement.” Democratic senator Edward Kennedy of Massachusetts called the legislation “a defining issue about the future of our nation and about the future of democracy, the future of liberty, and the future of the United States in leading the free world.”2
In retrospect, NCLB seems foreordained, because there were so many precedents for it in the states and in Congress in the previous decade. In the 1990s, elected officials of both parties came to accept as secular gospel the idea that testing and accountability would necessarily lead to better schools. Of course, testing was necessary to measure student academic performance and to determine whether it was moving forward, sliding backward, or standing still. At the time, few realized that the quality of the tests was crucial. Elected officials assumed the tests were good enough to do what they were supposed to do—measure student performance—and that a test is a test; they did not give much thought to such technical issues as validity or reliability. Everyone, it seemed, wanted “accountability.” By accountability, elected officials meant that they wanted the schools to measure whether students were learning, and they wanted rewards or punishments for those responsible.
The Death and Life of the Great American School System Page 12