Two decades ago Texas became ground zero for the accountability movement in public education. Now, after a revolt by teachers and parents who claim that High-stakes testing is ruining classroom instruction, the Legislature is poised to undo many of its own reforms. Does anyone have the right answer?
During his first run for the White House, George W. Bush called it the Texas Miracle. High-stakes testing in the public schools, along with other measures meant to hold teachers and principals “accountable” for the performance of their students, had closed the achievement gap between Anglo and minority students and boosted overall scores in reading and math. On the campaign trail, Bush touted the reforms—first passed by the Legislature in 1993, a year before he was elected governor—as a blueprint for the nation. And indeed, just a year after he arrived in Washington, the Texas model went nationwide when Bush signed No Child Left Behind into law, in January 2002, requiring all states to create their own testing programs. At the signing ceremony, Bush singled out Rod Paige, his secretary of education, whose success as the superintendent of the Houston Independent School District had provided the inspiration for the accountability movement in Texas.
After eleven years of this unprecedented experiment in American pedagogy, during which time student assessment grew into a $1.7 billion industry dominated by a handful of corporations, nobody is talking about miracles anymore. Not in Washington, where the Obama administration has been forced to grant waiver after waiver as NCLB’s ambitious 2014 deadline for states to reach “100 percent proficiency” in math and reading approaches. (In 2011, 48 percent of the nation’s schools failed to meet the law’s benchmarks.) And certainly not in Texas, where the Houston school district’s putative academic successes, including its astonishingly low dropout rate, have been debunked as statistical chicanery. Across the state, a long-simmering anti-testing movement has finally exploded into a full-fledged revolt. And it hasn’t happened only among teachers and administrators, who have argued for years that testing takes up too much time and energy. It has flared up in the demographic that animates public policy more than any other: suburban parents.
There was a time, difficult to remember today, when standardized testing meant putting aside class work for a single afternoon to take something like the California Achievement Test, which measured how well kids performed in math and reading in comparison with their peers around the country. Such tests were used as diagnostic tools, and there were no particular consequences for failing to do well. Those days are long gone. The accountability system Bush championed has continued to evolve, and the number and complexity of the tests have increased with each new legislative overhaul. Students in Texas are now tested a total of seventeen times in grades three through eight, and they must pass reading and math exams in grades three, five, and eight—or risk being held back. Schools that don’t perform well are subject to sanction by the state, including the dismissal of staff. Campuses that consistently lag behind can even be closed, a fate that befell Johnston High School, in Austin, and Sam Houston High School, in Houston, in 2008.
But it was the new requirements for high school students in the latest iteration of the tests, known as STAAR (State of Texas Assessments of Academic Readiness), that really set the woods on fire. Created by the Legislature during the 2007 and 2009 sessions and implemented last spring, STAAR replaced the four tests previously required for graduation with fifteen—more than any other state. Lawmakers also mandated that the new tests, known as end-of-course exams, would count for 15 percent of a student’s grade in each subject area. Further, kids who didn’t score high enough on the Algebra II and English III tests wouldn’t be eligible to attend any of the state’s public four-year universities. For middle-class parents, testing was suddenly more than just a nuisance—it was a threat to their children’s future.
If there was any doubt that testing advocates would be playing defense in the 2013 legislative session, it did not last past the first day, when Speaker of the House Joe Straus delivered his opening address. “The goal of education is not to teach children how to pass a test but to prepare them for life,” he said, adding, “To parents and educators concerned about excessive testing, the Texas House has heard you.”
A considerable number of angry mothers—and quite a few well-spoken kids—descended on the Capitol in February for a series of hearings on high-stakes testing convened by Dan Patrick, the Republican chair of the Senate Committee on Education. The February 19 hearing quickly turned into the morning of the long knives for testing advocates, including Walter Sherwood, an executive from Pearson, the world’s largest education company, which holds the state’s current five-year, $468 million testing contract. In the overflowing hearing room, Patrick wanted to know why kids had done so poorly on the first round of STAAR tests, taken last spring, when Pearson had promised that the questions would be tailored to the state curriculum. Failure rates for the English I writing and reading tests were 45 percent and 32 percent, respectively. “Either the teachers and the schools are doing a poor job of teaching the curriculum or you all are incorrect that these tests are accurate tests,” Patrick said. “Which is it?”
Never mind that Patrick had been a key player in the creation of the new tests, back in 2009. The political landscape has shifted considerably since then. At a national meeting of school administrators in January 2012, then–Texas education commissioner Robert Scott, who had been a staunch defender of testing, shocked the establishment when he said testing had become “a perversion of its original intent.” Tom Pauken, then the chair of the Texas Workforce Commission and a former head of the Texas Republican Party, piled on with a series of opinion pieces in which he referred to a “cult of educational testing” and argued that the accountability system was doing more harm than good. After nearly twenty years of high-stakes testing in Texas, Pauken and others pointed out, scores on the SAT hadn’t improved. A resolution opposing high-stakes testing has now been signed by more than 80 percent of the school boards in the state. Longtime observers of education policy are openly speculating that we are seeing the beginning of the end of the accountability movement, right here in the state where it was born. There is no question that the Legislature, which wraps up its current session on May 27, will roll back at least some of the accountability system in Texas. The question is, How far will it go? And how far should it?
Like Jihad and skateboarding and small furry animals, high-stakes testing has given rise to a new genre of YouTube video, a kind of inspirational training film meant to be viewed just before the testing season begins. Some are slickly produced, while others are clearly homemade, though they all tend to share some common tropes: students imitating rappers, teachers gamely chiming in, a dance beat pumping while kids chant “Rock this test!” and other mantras. Children are shown marching into class, poring over work sheets, learning “strategies” to beat the test makers, rallying in the gym, and so forth. The songs are upbeat and the kids, especially the third graders, are cute. But after watching a dozen of these clips, the relentless support-building becomes a little disturbing. You begin to feel as if you’ve fallen asleep in the first act of To Sir, With Love and awoken in some kind of Maoist reeducation camp. In “STAAR: Right Now,” a professionally produced video created by the Mesquite Independent School District, a series of elementary school children apply themselves with deadly earnest as an eighties-style rock anthem plays and increasingly ominous title cards flash across the screen. About two minutes in, a blond girl in a purple T-shirt is shown in an archetypal pose of anxiety: staring down at her work, supporting her chin with her hand. On the screen, we read, “Right now . . . a young girl is scared about passing the STAAR test. She’s worried.” Then a jarringly stark admonition appears across her face: “Does she need to study more? YES! Everything depends on it.” Everything? Really? We all know the global economy is more competitive than it once was. But this is an eight-year-old girl trying to pass the third grade, not an astrophysicist plotting a launch window.
Of course, the stakes are high not just for kids but for adults too. Teachers whose students fail to improve—or don’t improve sufficiently—can find themselves out of a job. Students at poor-performing schools can transfer to better schools in their district—and take their state funding with them. Perhaps inevitably, the pressure to turn failing schools around quickly has led to some embarrassing scandals. Teachers and administrators have been caught changing scores on tests, encouraging low-performing kids to stay home on test day, and falsely labeling kids as disabled or non-English-speaking so they can take easier tests. But none of these stories compare to the scheme that went on for years in the El Paso Independent School District. Last fall, superintendent Lorenzo Garcia was sentenced to federal prison for contracting fraud, but investigators also discovered that he had used a variety of means to force out dozens of marginal students in an effort to boost test scores—and collect testing-related bonus pay. A system that was designed to decrease the dropout rate was, in fact, having the opposite effect. Across the country, meanwhile, teacher morale has sunk to its lowest point in 25 years, according to the annual MetLife Survey of the American Teacher.
How did it come to this? Rumblings of a coming invasion of “market-based” reforms for schools had been building since the mid-eighties, but the framework for the modern accountability system was a report produced in 1993 by the Educational Economic Policy Center, a state agency tasked with improving troubled public schools. One of the report’s main architects, a Dallas lawyer and school board member named Sandy Kress, explained its findings that April in an opinion piece in the Dallas Morning News. Titled “Schools Must Be Held Accountable for Performance,” the editorial argued that research “provides clear evidence that even among schools with high numbers of low-income children there are great differences in achievement scores. This means that something is happening in some schools to improve student performance that is not happening in others—that a child’s background is not the only explanation. It means that schools and the people running them can and should be held responsible for student results.” Lieutenant Governor Bob Bullock asked Kress to oversee the creation of a new assessment system for the state’s public schools. The program, passed by the Legislature in 1993, required schools to disaggregate standardized testing data for the first time—breaking down how well minority, poor, and disabled students were performing as individual groups—and began ranking campuses based on their test scores and graduation rates.
The shaming effect of public rankings had administrators scrambling to improve performance. As the exams became more and more difficult and the stakes grew higher and higher, the school year in Texas became more and more centered on testing. It wasn’t just the test days themselves. Schools also employed so-called benchmark tests, to see how students were progressing throughout the school year. In the weeks leading up to the actual tests, they pulled kids from “nonessential” classes like art and gym to do endless drills and work sheets, a dreaded routine that came to be known as “drill and kill.” There were strategy sessions in which students learned how to think like a test maker. One of the first tricks: on a multiple-choice question, begin by eliminating the obviously wrong answer; that way you’ll be more likely to accidentally get the right answer.
Accountability wasn’t cheap. There was software to buy to help teachers tailor their lesson plans to the test and track how well kids were doing. Districts also hired employees who did nothing but collect and interpret test data; at the Texas Education Agency, nearly one hundred positions were eventually created to fill a new accountability department. Tutoring became a growth industry in Texas. After the passage of NCLB, which requires districts to provide extra help for failing students, it ballooned into a $4 billion industry nationwide.
When Bush left for Washington, Kress became his senior adviser for education. He found himself in great demand as the testing market metastasized, and he began consulting and lobbying for the England-based company Pearson. He returned to Texas, landing at the Austin office of the prestigious law firm Akin, Gump, Strauss, Hauer, and Feld, where he became Pearson’s point man in Texas.
When I called Kress for this story, he agreed to talk to me on one condition: nothing we discussed would be on the record. This may sound like an odd request from a man who has been the public face of high-stakes testing for twenty years, but in recent months the political atmosphere at the Capitol has grown increasingly poisonous for him. In the course of a two-hour interview in his office high above downtown, the 63-year-old Kress seemed very much like a man under siege. With a slight build and a full head of brown hair, he was dressed casually, in a blue-checked shirt and loafers. The walls were covered with photographs, commendations, and other mementos of happier times: Here he was with Governor Bush when Kress was his brash education adviser. Here he was with Senator Ted Kennedy, whose support of NCLB ensured that the bill would pass with bipartisan support.
Kress has lobbied for Pearson for more than a decade, but until recently this link seems to have done little to undermine his legitimacy as a pioneer in education reform. It probably hasn’t hurt that Kress doesn’t always identify himself as a Pearson lobbyist when he offers his opinion online or in print—including, for example, two op-eds in the Austin American-Statesman defending high-stakes testing in the months before this year’s session of the Legislature convened. As anti-testing sentiment has spread, however, Kress has been blindsided by the new and growing notion that the accountability movement is being driven by greed—specifically Pearson’s greed. Shortly before he resigned as education commissioner, Robert Scott warned that assessment and accountability had grown into something akin to a “military-industrial complex,” a kind of public policy juggernaut with its own internal momentum. It’s an argument that has caught fire with grassroots opponents in Texas, and the inevitable result is that Kress’s brand has been damaged along with Pearson’s.
Things started to turn sour for Kress personally last summer, right around the time an Austin-based Democratic consultant named Jason Stanford penned a widely disseminated anti-testing screed. Stanford wrote that he had asked his third-grade son what he had learned that year and heard this back: “I learned strategies. How to focus and . . . I forget the other one. Oh, yeah, practicing doing tests.” Stanford criticized Kress directly, noting that Kress’s kids went to private school and didn’t have to take the tests he had helped foist on everybody else. Kress responded in high dudgeon, pointing out that he had been an advocate for accountability for ten years before he went to work for Pearson. Later, on an education blog, Kress elaborated on his response, rejecting the notion that high-stakes testing was turning Stanford’s third grader into a professional test taker. “I can hear the cry already: ‘they’ set such high stakes for the test, ‘they’ have ‘made us’ do these stupid things,” he wrote. “No! Nonsense! The tests should have consequences. But drill and kill is no solution. Good teaching to the standards is the only solution that works, and I refuse to be identified with the crap that is done instead.” None of this, he seemed to be saying, was his fault.
If you believe that the Legislature overreached when it created the new STAAR tests, however, a good case can be made that this mistake was, at least in part, Kress’s fault. The switch to end-of-course exams—passed by the Legislature in 2007 but implemented only last year—was a reform meant to mollify teachers who thought their kids were not being tested on the specific curriculum they were required to teach. But from the beginning, there was debate over how many tests students should have to pass, with lawmakers in the House favoring as few as five. Kress and others, however, argued that the problem with the accountability system was a lack of rigor in high school. Specifically, the exit exams under the old system, known as TAKS, were thought to be too easy, allowing students to graduate without ever moving beyond the tenth-grade level in subjects like math, reading, science, and social studies. The answer, they felt, was more testing.
The push for more rigor found a champion in then-senator Florence Shapiro, a Republican from Plano who served as the chair of the Senate Committee on Education, and the result was a bill that required students to pass fifteen end-of-course exams to graduate. Pearson, which has held the state’s contract for more than a decade, was hired to create the new tests.
In the years that followed, Kress kept up a steady drumbeat for increasing the rigor of high school testing. In 2007 Governor Rick Perry appointed him to chair the Commission for a College Ready Texas. Kress and his fellow commissioners toured the state talking to presidents of community colleges and universities as well as employers. What they found was sobering: Business owners complained that high school graduates couldn’t understand instruction manuals or comprehend a simple memorandum, much less write one. Forty percent of kids entering community college needed remedial classes.
The next year, Perry appointed Kress to the Select Committee on Public School Accountability, which was charged with further overhauling TAKS. The committee heard testimony from teachers, parents, and administrators, many of whom urged members to undo some of the harsher aspects of the state’s accountability system. But Kress held firm, later helping to fight off another attempt by House members, in the 2009 session, to limit the number of tests needed to graduate.
The result is that Texas is now an outlier when it comes to exit-level exams. The only state that comes close to requiring so many tests is Virginia, which mandates nine at the most. Twenty-five states require no exit exams at all. Of the other states that do, seventeen require three or fewer. “It’s classic Texas,” said David Anthony, the CEO of Raise Your Hand Texas, a nonprofit that supports greater investment in education. “If testing is good, then more testing will be even better.” Perry’s decision to appoint Kress, who had been lobbying for Pearson for years, to a committee charged with overhauling the testing regime for public schools was classic Texas too. Nobody doubts Kress’s deep knowledge of the field, but would a lobbyist for a testing company ever recommend that kids take fewer tests?
In retrospect, the revolt against the accountability movement could have been predicted two years ago. Traditionally, any ratcheting up of accountability measures has been accompanied by an increase in funding; this time, the opposite happened. During the 2011 session, when the decision was made to cut an unprecedented $5.4 billion from public education to deal with the recession-induced shortfall, some legislators tried once again to modify STAAR, which by then was almost ready to be implemented. House Public Education Committee chair Rob Eissler, a Republican from the Houston suburbs, suggested reducing the number of exit exams and eliminating the requirement that test scores count for 15 percent of a student’s grade. Educators pointed out that funds to support remedial instruction for children who failed STAAR exams were being cut, yet Pearson’s contract, worth more than $90 million per year, was untouched. But accountability advocates, led by Shapiro and backed by Bill Hammond, the president and CEO of the Texas Association of Business, dug in their heels. Eissler’s effort failed.
This session, Hammond has emerged as testing’s staunchest defender. On a morning in February when the Legislature hosted a pair of key hearings on testing, he paid for a plane to circle the Capitol pulling a banner with a decidedly unsubtle message. “Is 37 percent correct on Algebra too hard?” the banner read, alluding to the low bar the Texas Education Agency had set for “passing” the first round of STAAR tests. This aggressive tone is nothing new for Hammond, who has been one of the Capitol’s most reliable accountability advocates and who famously accused Robert Scott of being a “cheerleader for mediocrity” after the commissioner backtracked on testing. What is new this session is that Hammond seems to be increasingly isolated on the issue. He’s even begun to take sharp criticism about his motives, from legislators like San Antonio representative Mike Villarreal, who pointed out that since Pearson is a member of the Texas Association of Business, Hammond arguably should be considered a de facto lobbyist for Pearson. Of course, Hammond lobbies on behalf of the business community in general, which has a vested interest in maintaining a good public school system, but it’s hard to think of a company that the TAB has done more for this session than Pearson. When I asked him about the accusation, Hammond pushed back. “I take great exception to anyone who would impugn my integrity,” he told me. “I have a thirty-year record of being for education reform that goes back to when I was a freshman in the Texas House.”
A little less time spent talking to employers and college presidents and a little more time talking to suburban parents might have headed off the coming storm too. For all the changes wrought by the school accountability movement over the past generation, life for kids at high-performing public schools didn’t really feel that different, at least until STAAR came along. Those students have traditionally obsessed over college entrance exams, not the state-administered tests for which struggling kids and schools spend so much time preparing. The end-of-course exams under STAAR changed all of that. Now a student’s grade point average—and his or her chance of getting into college—could be affected by the state tests. A small group of Austin parents calling themselves Texans Advocating for Meaningful Student Assessment (TAMSA) began organizing in 2011, focusing, at first, on getting rid of the 15 percent rule. The group received a boost in January 2012 when Susan Kellner, the onetime board president of the Spring Branch Independent School District, in west Houston, joined the cause. Kellner’s husband, Larry, a former chairman and CEO of Continental Airlines and a major Republican donor, had served on the Select Committee on Public School Accountability with Kress, and she had accompanied him to hearings held around the state. Part of west Houston is in Dan Patrick’s district, and the Kellners are two of his most prominent constituents. TAMSA certainly seems to have Patrick’s ear: the powerful senator eventually filed a testing reform bill, Senate Bill 1724, that would dramatically reduce the number of tests students must pass to graduate.
Susan Kellner, whose four kids have attended public schools, looks like Sandra Bullock—not the awkward America’s sweetheart version from Miss Congeniality but the one from The Blind Side, who nobody wants to mess with and then regrets it when they do. On a sunny Austin day in early March, during a break between hearings, Kellner told me about her aha moment. Years ago, when her youngest kids were in the first and third grades, she managed to get an author of science books for children to agree to speak for free to both classes. At her third grader’s school, located in the Kellners’ affluent neighborhood, the teacher was delighted. But Kellner’s first grader attended a school in a less affluent neighborhood that offered a dual-language curriculum. “Oh no,” her teacher replied. “That’s the week before TAKS, and we just can’t free up the time.” Kellner was flummoxed. “I thought if any children need exposure to this enriched, fun way to learn about science, it’s the children from poverty or minority backgrounds, not the wealthy white children,” she recalled. Kellner told me several times that she is not anti-testing—she believes tests are useful for diagnostic purposes. But unlike David Anthony and Robert Scott, who both feel STAAR is a good testing regime and should be kept, at least as a diagnostic tool, Kellner advocates chucking the whole system and instead using nationally administered standardized tests like the ACT, which offers not only a college entrance exam but also a variety of tests for younger students to measure their progress toward college readiness.
The way policy gets made in Texas was another eye-opener for Kellner. “First of all, we couldn’t believe that Sandy Kress was invited to participate in those interim hearings,” she said. “Talk about the fox guarding the henhouse.” From the beginning, Kellner said, Kress butted heads (albeit politely) with her husband and other committee members, who he felt were trying to water down the accountability system. Going into the 2009 session, Kellner recalled, it looked as though testing critics would at least get a reduction in the fifteen end-of-course exams required for graduation under STAAR. The House wanted to go down to three; the Senate proposed going down to ten. “But there was never any talk of still having to pass fifteen of these tests to graduate,” Kellner recalled. That all happened in closed-door negotiations on testing reform in the final days of the session. “Governor Perry threatened to veto anything that decreased the number of high-stakes tests,” she said, and the members caved. This session, Kellner said, the dynamic has been much the same. “If it was up to the House, I really believe there would be no high-stakes testing at all. But what we keep hearing from members is ‘I’d like to do more, but if we gut high-stakes testing, Perry will veto it.’ ”
When I mentioned Hammond and Kress’s contention that poor results—only 23 percent of high school graduates are considered “career or college ready”—were not an indictment of high-stakes testing in general, just of the particular tests the state has been administering, she scoffed. “Oh, it’s the wrong test,” she said. “We’ve been hearing that for years—TABS, TEAMS, TAAS, TAKS—and now they want to reinvent the wheel again. When are you gonna have the right test?”
Amid all the havoc that high-stakes testing has wreaked, there’s a fundamental question that can get overlooked: Does it actually work? Are public schools doing a better job teaching our kids than they used to? Prior to the shift to STAAR, scores on state-administered standardized tests had gradually been trending upward in Texas. But such trends are widely regarded as suspect by education researchers because test scores, applauded for introducing an element of objectivity into the effort to assess student performance, in reality do no such thing. Once the tests have been taken, it’s up to the Texas Education Agency to set the “cut rate”—the number that represents a passing score. Administrators can, and do, adjust that number to reach “passing” rates that they consider to be politically acceptable. You can’t flunk everybody, after all. (Hammond’s “37 percent” banner nicely illustrated this phenomenon. How on earth could that be considered a passing score?) New York mayor Michael Bloomberg, another staunch advocate of accountability, was famously left with egg on his face when a much-trumpeted decline in the achievement gap in his city disappeared in 2010, after rampant manipulation of cut rates was exposed. In any case, most experts agree that students simply get better at taking a test after they and their teachers have had a lot of practice with it. Whether they are learning anything beyond test-taking is another question.
Defenders of high-stakes testing in Texas point to improving scores on the U.S. Department of Education’s National Assessment of Educational Progress, a test that has been given annually for the past twenty years or so to students in the fourth and eighth grades. Cheating on the NAEP appears to be rare, and unlike the TAKS or STAAR tests, Kress and others have argued, the NAEP cannot be gamed. Overall results for Texas show modest improvement over the past two decades—not much better or worse than any other state. Among certain subgroups in certain subjects, however, there have been encouraging signs. In math, black and Hispanic students are testing roughly four grade levels ahead of where they were in 1990. The gap between minority and Anglo students in Texas has closed, though improvement in reading scores has been more modest.
But are NAEP results really the gold standard some testing advocates claim? The Department of Education selects a random sample of fourth and eighth graders from each state to take the NAEP, but state officials can exempt some students from the testing pool. The idea is to exclude kids with developmental disabilities or those who don’t know enough English to pass the test. The problem is that there are no firm criteria for making this determination. Kress gave me a report highlighting how well Texas has fared in recent years on the NAEP in comparison with the other mega-states: California, New York, Florida, and Illinois. Yet the report also shows that Texas exempts students at a much higher rate than any of these states (on the fourth-grade reading test, for example, Texas’s exemption rate was twice the national average and six times higher than California’s). Exempting students from the test is meant to be done as a last resort; other remedies, such as longer testing times, one-on-one testing, and simpler tests for some students, are supposed to be used first.
The governing board that oversees the NAEP recognizes that test results are being skewed and has tried to fix the problem, but the Department of Education has resisted. As with any testing regime, there are political considerations: if fewer marginal students are exempted, NAEP scores—a key benchmark used to demonstrate the putative success of No Child Left Behind—will go down. Paul E. Peterson, an education scholar at Harvard, observed that students in general have been doing better on the NAEP but speculated that much of the gain is likely illusory, the result of pressure from various sources to make the tests easier. Tellingly, he noted, American students show much more modest gains over time on well-regarded international tests. Our position relative to other industrialized nations has remained largely unchanged since the sixties: middle of the pack.
The Bible for opponents of high-stakes testing is a 2010 book called The Death and Life of the Great American School System: How Testing and Choice Are Undermining Education, by Diane Ravitch, perhaps the nation’s preeminent education historian. Ravitch, who grew up in Texas and attended Houston public schools, was once an advocate of both high-stakes testing and charter schools. She served as the assistant secretary of education under President George H. W. Bush and was later appointed by President Bill Clinton to head the National Assessment Governing Board. Her embrace of market-based education reforms landed her a fellowship at the Hoover Institution, the conservative Stanford University think tank that has produced much of the intellectual underpinning of the accountability movement. “Like everyone else,” she told me during a stop in Austin in February, “I was drawn to the idea that schools might benefit from a business sensibility, that we should set goals and then reward high performers and punish low performers.”
Over time, however, she became disenchanted. As a historian, Ravitch has spent her career cataloging the various fads and trends that have swept through public education over the years. Eventually she came to see the accountability movement in the same light: a vast experiment promising a panacea with little evidence in support of its claims and much harm done in the process.
Ravitch, who is 74, has in many ways become the anti-Kress, a Texan on the national stage who rejects the Texas-born solution that has been exported to the nation’s schools. Like Kress, she is tireless. She has spent the past three years on a kind of never-ending book tour, warning about the false promises of a movement she once supported. “Whenever I come back to Texas, I tell people, ‘This idea was born here, and this is where we should put an end to it,’ ” she told me. She often debates policy on obscure websites with researchers like her former Hoover Institution colleague Eric Hanushek, who testified at a recent hearing in Austin. Hanushek told legislators that simply removing the least effective 8 percent of teachers would make our education system as good as Finland’s, which is to say, as good as any in the world. This is an oft-repeated talking point for accountability advocates; Bill Hammond mentioned it in an Austin American-Statesman op-ed on March 11, in which he argued that the ability to fire teachers at will would make a bigger difference than, say, increased funding.
But that argument isn’t a particularly compelling one in Texas. The teachers’ unions, so powerful in places like New York and Pennsylvania, hold little sway here, where teachers cannot bargain collectively or go on strike. There is no tenure for the vast majority of teachers in Texas, most of whom have one-year contracts that school boards can decline to renew “for good cause,” which could mean failing to meet any stipulation in the contract, including student performance on standardized tests. Teacher turnover is high in any case, which poses a conundrum for those who feel teachers are the problem: After the “bad” teachers leave, how do you find and keep the “good” ones?
Likewise, Finland, where students do not take a standardized test until after graduation and where class grades are not even recorded until fifth grade, is a curious example of success for defenders of the accountability movement in Texas to seize upon. Finland doesn’t really have an accountability system. What it does have is a truly professional teacher corps in which salaries are relatively high, turnover is low, and advanced degrees are the norm. Finland also differs dramatically from Texas in another respect: its child poverty rate is 5 percent; ours is 26. (In fact, 60 percent of our public school kids qualify for free or reduced-price lunches, meaning they are either officially poor or not far from it.) “The number one determinant of how well kids will do in school is socioeconomic background,” Ravitch told me. “It’s not how good your teacher is or which school you go to.” Ravitch makes a convincing case that those pining for a lost golden era of American education are misremembering. Sixty years ago, black and Hispanic kids weren’t allowed to attend public schools—or at least, not real ones—and most didn’t even go to high school. Kids with disabilities were excluded as well, and there were far fewer recent immigrants enrolled. Comparing that system with the one we have today makes no sense.
Many observers trace the origin of the current wave of school reform to 1983, when the Reagan administration published “A Nation at Risk,” a sober assessment of declining student performance that warned of a “rising tide of mediocrity” threatening America’s place as the leader of the free world. There was no mention of testing or accountability in the report, which focused instead on improving the curriculum used in public schools. An early attempt at drafting a new national history curriculum foundered, however, when conservatives detected too much emphasis on oppressed groups and national failures and not enough on America’s triumphs. Curriculum reform had fallen victim to the culture wars, and thereafter most politicians wanted nothing to do with it. What emerged instead were market-based reforms and data-driven analysis. Ravitch believes that the accountability movement began with good intentions but has now evolved into a useful tool for those who would like to weaken, or eventually replace, the public education system. “I heard the achievement gap spoken about in such cynical terms among some of my conservative colleagues,” she told me. “What they really want to do is crush the teachers’ unions and privatize as much of the system as possible.” No Child Left Behind hasn’t had that effect thus far, but it certainly could be used that way if officials in Washington and local school boards wanted to do so: under the law, schools that fail to make “adequate yearly progress” for three years in a row can be shut down and reconstituted as privately run charter schools. Ravitch doesn’t believe Sandy Kress set out to undermine public education, and neither does anyone else who really knows him. “I’m sure he believes in what he is doing, but he is still talking about how testing can help in theory,” she said. “What the rest of us are dealing with is reality.”
Are we really seeing the end of the accountability movement in Texas? Both House Bill 5, by Republican House Public Education Committee chair Jimmie Don Aycock, and Senate Bill 1724, by Dan Patrick, seem certain to pass, which means that the hated 15 percent rule will be done away with and the fifteen tests required to graduate from high school will be reduced to perhaps as few as five, presumably making Pearson’s next contract much more modest. But life in elementary and middle school is not likely to change much even if the bills become law. None of the reforms under serious consideration would do away with the requirement that these kids be tested every year, for example. Nor can they: federal law requires annual testing in reading and math for all students. States that don’t comply lose millions in federal funding. There will still be drill and kill, endless strategy sessions, and harried and unhappy teachers. “When I first came down here, I was so idealistic,” Kellner told me at the Capitol. “I thought they’d see it wasn’t working, and we could get rid of the whole thing. Now I know we have to take what we can get.”
A day after I met with Kress, I received a short email from him. He had changed his mind about giving me a quote. Here is what he decided to say: “Because weakened accountability will most hurt low-income kids and kids of color, everyone mentioned in this story should be called back to the scene in ten years and held to account for the resulting student outcomes.” A reader could be forgiven for thinking something along the lines of “Could it get much worse?” The answer could very well be yes. Or, if accountability does fade away, perhaps some other idea will take hold in the meantime: a renewed focus on crafting a rigorous curriculum, as Ravitch recommends, or an effort to attract and retain talented teachers. Bringing Finland to Texas seems like a distant dream now, but who knows? A lot can happen in ten years.