Kalamazoo, Michigan, provides a powerful example of the hype-disillusionment cycle characteristic of teacher merit pay plans. In 1974, the American School Board Journal published a rapturous series of articles about the district’s new superintendent, William Coats, who had never before worked in public school administration (he had been a college professor). Coats had proposed a merit pay system in which every teacher would be graded according to her students’ standardized test scores. Her classroom practice would be observed by five to fifteen peer teachers, by her principal, and by pupils, all of whom would produce written reports on her performance. The teacher would also write a self-assessment. Administrators, too, would be evaluated by their colleagues, by the teachers who reported to them, and by students and parents.
Just a year after implementation, the Journal reported that this system had transformed the Kalamazoo public schools. “Student achievement is up—significantly so. Racial violence, which for years has plagued Kalamazoo’s schools, has almost disappeared. Board members smile more pleasantly and more frequently, they report, confident for the first time in years that most taxpayer dollars are not wasted in city schools.”
In reality, superintendent Coats and his merit system lasted less than three years in Kalamazoo. The plan attracted immediate backlash from every corner, including from administrators exhausted by the overwhelming number of employee reports they had to write. Although test scores improved—in reality, only modestly—parents did not celebrate and instead complained about classroom time spent prepping for multiple-choice tests now that teachers were being graded according to their students’ scores. As for the decrease in racial violence, it had little or nothing to do with merit pay. In 1971 the courts had ordered Kalamazoo to desegregate its schools, which put black and white children into contact with one another at the elementary school level, so that by the time they entered the city’s two high schools, racial tensions were less extreme. The district renovated classrooms in formerly all-black schools, hired more black teachers, and actively recruited black children into advanced courses. Coats had been hired by a majority-white school board that opposed the court desegregation order. He claimed his mechanistic teacher evaluation scheme had produced the district’s student achievement gains. But principals disagreed. “Now the schools are free to focus on performance and academics, since desegregation has eliminated the racial conflicts and the cost of controlling those conflicts,” a Kalamazoo junior high school principal told the U.S. Commission on Civil Rights. “In my school we need fewer crisis staff and it is now a quiet school.”
All over the country, heavy paperwork burdens turned principals into skeptics of politically imposed merit pay plans—much as the dissident principal Alexander Fichlander had complained, in New York City in 1920, that the city’s complex A–D evaluation scheme sucked up time while doing little to help teachers improve. In Texas, a 1984 guidebook advising principals on how to hold teachers accountable suggested an evaluation rubric with sixty-three categories, each requiring a rating on a scale of one to five. Some of the items contradicted one another. Were teachers ignoring “attention-grabbing” behavior from students? That was a good thing. Yet another category called for teachers to quickly “negatively reinforce” bad behavior.
There were several merit pay plans that were popular with teachers, and they had one major feature in common: Bonuses were available to every high-performing teacher in a school district, regardless of the grade level or subject they taught. That was rarely the design of merit pay programs launched in the wake of A Nation at Risk, which tended to be funded at such low levels that bonuses were available only to teachers of some subjects and grades: just 15 percent of teachers in Tennessee under Governor Lamar Alexander’s widely celebrated plan, and 10 percent of teachers in Florida. When so few teachers could benefit from merit pay, the bonuses were seen as divisive, undercutting morale.
Gera Summerford, today the president of the Tennessee Education Association, the state teachers union, was a first-year middle school math teacher in Nashville when the merit pay program rolled out in 1982. “I was very idealistic,” she recalled. “I felt like it was an opportunity for me to demonstrate that I was a more effective teacher than maybe some of the ones who had been teaching for a long time. But when I went through the process, I was very disillusioned.”
Summerford was supposed to be evaluated by three people, including a peer with experience in her subject area. But the district sent a business teacher, not a math teacher, to observe her classroom, and she found this unhelpful. The system applied to both teachers and administrators, and it created three levels of recognition, with bonuses between $1,000 and $7,000. Teachers soon noticed that almost every principal who applied for the extra pay was quickly bumped up through all three levels, while just 13 percent of teachers achieved Level 2 or 3. That created resentment. The most popular part of the system was overtime pay offered to teachers who supervised afterschool programs or provided extra tutoring. “That really did help kids,” Summerford said. But ultimately the relatively small bonuses tied to capricious classroom observations seemed almost absurd, “an insultingly small amount of money” distributed through a labor-intensive process yet in an unpredictable way. Within a decade, the system was phased out, in part because of continued NEA opposition, and in part because it had been financed by an unpopular increase in the state sales tax.
All this foment around merit pay occurred at a time when the average American teacher earned just $23,500 per year ($47,400 in today’s dollars), less than a mail carrier and half that of the average lawyer or accountant. An alternative corner of the education reform movement tried to send the message that raising base salaries—by 25 percent, the Carnegie Foundation recommended, or to an average of $50,000, according to RAND—would do much more to increase the profession’s prestige.
But influential business leaders were eager proponents of numbers-driven merit pay for teachers. Ross Perot, for example, pushed Dallas to implement a plan to use test scores alone to evaluate teachers and distribute pay increases. So it was ironic that private industry had, by the 1980s, mostly turned away from efforts to pay white-collar workers according to strict productivity measures, finding that such formal evaluation programs were too expensive and time-consuming to create and implement. Research showed that companies with merit pay schemes did not perform better financially than did organizations without it, nor were their employees happier. Instead, management gurus recommended that workers be judged primarily by the holistic standards of individual supervisors.
It is important to note that in public education, teachers unions have had a long history of fighting attempts to institute subjective evaluation systems; at the most basic level, New York City teachers had gone on their massive strike in 1968 to deny an administrator, Rhody McCoy, the right to evaluate his teachers and staff his schools as he saw fit. In the 1980s both the AFT and the NEA responded to calls for merit pay by saying they would support it only when it was based on “objective” measurements, such as teachers’ scores on tests of subject matter knowledge. Al Shanker said, “I always was and still am against any merit pay notion that would be based on the idea that some principal walks into your room and gives a judgment as to who is a better teacher.” Some accountability reformers actually shared Shanker’s skepticism. Chester Finn, a longtime moderate Republican education reform guru who advised Lamar Alexander on teacher career ladders, described to me the grim status quo in classroom observation: “The principals were often former gym teachers, and had almost never been trained to be sophisticated overseers of teacher quality or performance. In the absence of quantitative data, a principal would be a perfunctory and ill-trained observer, and either liked or didn’t like what he saw. It was awfully easy for the principal to then make decisions based on ‘Who do I like? Who’s my cousin? Who’s my girlfriend?’ or a variety of other factors that don’t have to do with classroom effectiveness.”
When unions brought t
his suspicion of principals to the negotiation table, the results in some instances were even more complex and expensive evaluation systems, like Tennessee’s, in which principals’ ratings were cross-checked and sometimes overruled by teams of observers from outside each school, a plan Shanker approved. In the end, nearly all the merit pay and career ladder plans of the Nation at Risk era—underfunded, unpopular, and overly bureaucratic—were defunct by the end of the 1980s. Supposedly more objective measures of teachers based on student test scores would emerge in the national conversation in the late 1990s, in large part because both union leaders and some reformers had for so long refused to trust principals to evaluate their staffs.
While merit pay failed in the 1980s, in another respect the Reagan administration achieved its education goals by displacing the previous era’s commitment to school desegregation.
On October 8, 1984, in the thick of his reelection campaign, President Reagan appeared in front of a majority-white audience in Charlotte, North Carolina. He claimed the city’s storied busing program “takes innocent children out of the neighborhood school and makes them pawns in a social experiment that nobody wants. And we’ve found out that it failed.” To Reagan’s surprise, he was greeted with stony silence, not cheers.
In response to a 1971 Supreme Court ruling, the Charlotte-Mecklenburg school district had developed a desegregation plan that paired kindergarten through third-grade schools in majority-white neighborhoods with fourth-grade through sixth-grade schools in majority-black neighborhoods. Each cohort of children was bused for half of their elementary school career, and no school in the district could be more than 50 percent black. What Reagan’s speechwriters did not realize was that the plan was popular and academically effective. Students of all races demonstrated steady test score gains in Charlotte after desegregation. A greater proportion of the district’s financial resources and experienced teachers moved to schools in low-income neighborhoods, which gained powerful advocates in the parents of middle-class and white children bused in from more affluent areas. The day after Reagan’s comments, the Charlotte Observer editorialized that the district’s integration program was, in fact, the region’s “proudest achievement.”
It was an achievement the Reagan administration actively worked to undermine. Early leaders in the teacher accountability movement were often simultaneously engaged in efforts to defund and delegitimize desegregation. In 1980, under President Carter, the Department of Justice filed twenty-two school desegregation suits. The following year, under President Reagan, it pursued just ten desegregation cases. In 1984 Secretary Bell spent $1 million of his discretionary budget to seed fifty-one new teacher merit pay plans across the country, even as the Reagan administration denied tens of millions of dollars of previously promised funding to school districts, such as Chicago’s, which were struggling to implement desegregation orders. Bell was even on the record in support of a federal law or constitutional amendment to forbid busing as an integration strategy.
In 1981 Reagan had appointed Robert Potter, a onetime anti-busing activist, to the U.S. District Court in western North Carolina. In September 1999, Potter ruled in favor of a Charlotte father who sued the school district to end busing, arguing that it discriminated against his biracial daughter, who was white and Latina. What happened next was tragic. According to research from the labor economist C. Kirabo Jackson, as Charlotte schools reverted to the demographics of their surrounding neighborhoods between 2002 and 2005, schools that became predominantly black suffered a loss of high-quality teachers as measured by growth in students’ test scores, teachers’ years of experience, and scores on teacher certification tests. Though black teachers were more likely than white teachers to remain in majority-black schools, those black teachers who left had been more effective than those who remained—especially at increasing the achievement of black students. Jackson estimated that this shift in the distribution of effective teachers could explain up to 7.5 percent of the achievement gap between black and white children in Charlotte. The movement of experienced teachers away from high-poverty schools has often been cited as proof of white teachers’ racist attitudes. And while discrimination certainly accounts for some of the movement, the fact is that many effective nonwhite teachers, too, seem to prefer working in integrated or middle-class settings. Part of the explanation is that high-poverty, minority-majority schools are more likely to experience administrative turnover and inept management, which erode teacher job satisfaction over the long term.
There is a wealth of other evidence that integration can boost student achievement. A second study of Charlotte, from economists Stephen Billings, David Deming, and Jonah Rockoff, found that between 2002, when schools resegregated, and 2011 the math achievement gap between the races widened. Young minority males assigned to Charlotte schools that experienced an influx of nonwhite students were more likely to be arrested and imprisoned than their demographically identical neighbors who remained in more integrated schools. In a separate paper, Byron Lutz of the Federal Reserve demonstrated that the end of court-ordered busing in northern school districts led to an increase in black high school dropout rates. (Interestingly, dropout rates did not change in southern districts that were released from busing orders. This could be because northern districts interpreted desegregation orders more faithfully than southern ones did, so the impact of lifting the court orders was more severe. Or it could be because some southern cities, like Nashville and Charlotte, pledged to accompany desegregation by increased investment in majority-black schools.)
One of the most compelling studies of the relationship between school integration and student achievement was conducted by Heather Schwartz of the Century Foundation. She examined Montgomery County, Maryland, where a lottery determines in which public housing development a low-income family lives. The children of families living in public housing within integrated school zones—where poor, typically nonwhite children were less than half of the student body at their schools—experienced an additional 8 points of math gains and 4 points of reading gains on 100-point tests, compared to demographically identical peers placed by lottery in schools where the vast majority of students were poor. In 1980 American school integration reached its height, with 37 percent of black children nationwide attending majority-white schools. After Reagan’s intervention, that progress eroded. By 2000, just 28 percent of black children were in majority-white schools, and 40 percent of black and Latino students attended deeply segregated schools, where 90 to 100 percent of the student body was poor and nonwhite.
The teacher accountability agenda that has emerged over the last two decades—stricter evaluation systems, merit pay, the weakening of teacher tenure, and the creation of alternative pathways into the classroom, like Teach for America—is often talked about as a sort of next step in school reform, because integration failed. In her 2011 book A Chance to Make History, TFA founder Wendy Kopp wrote, “In the sixties and seventies we committed to desegregate schools in order to ensure that all of our nation’s children have access to an equal education. Unfortunately, though, poor and minority students continued to lag academically.”*4 The conclusion that desegregation did not work is not fair, though—because the United States did not, in fact, commit to integration. In 1974 the Supreme Court ruled in Milliken v. Bradley that majority-white northern school districts had no responsibility to cooperate with inner-city schools toward the goal of integration, even in regions where affluent all-white school districts were just a few minutes away from urban neighborhoods ravaged by poverty. Desegregation was never widely implemented outside the South, and where it was implemented, as in Charlotte or Montgomery County, it often succeeded in raising student achievement to a similar or greater degree than did later teacher accountability reforms. Today there is still a demand for integrated schools. In cities like Boston and Hartford, tens of thousands of parents have their children on waitlists to be bused to higher-quality suburban schools. From Brooklyn to Atlanta to Los Angeles, a small gr
oup of socioeconomically diverse charter schools has proven enormously popular with families. So it is unfortunate that these two strains of American education reform, integration and teacher accountability, rarely work in tandem.
Since 1980 the federal government has done almost nothing to encourage local school districts to create racially and socioeconomically mixed schools, even as billions of dollars are sent to states and districts that agree to tie teacher pay and evaluation to student test scores and to open new charter schools, most of which are as racially and socioeconomically homogeneous as the schools the civil rights crusaders fought to reform.
When “education governor” Bill Clinton became president, he sought to push the standards and accountability movement further and faster. Through two pieces of 1994 legislation, the Improving America’s Schools Act and Goals 2000, Washington required states to adopt new curriculum standards and tests in order to receive Title I money. Clinton hoped to address one of Title I’s long-standing flaws: that states and school districts lacked the expertise to create high-quality curricular materials. He tried to establish a National Education Standards and Improvement Council where researchers would use best practices to develop standards, textbooks, and tests that states and local schools could choose to adopt. But in 1995 the GOP-controlled Congress withdrew support for the program, and it never launched. By the late 1990s, education reformers who had applauded A Nation at Risk were hugely frustrated at the continued lack of oversight of failing schools. States had new tests, but if scores remained low year after year, there were no consequences.
One of those frustrated reformers was Kati Haycock. She began her career in the 1970s, managing affirmative action programs and student outreach for the University of California system. That experience led her to conclude that college was too late to close achievement gaps between white and minority children, and that the real work was in K—12 school reform. In 1990 Haycock quit her job as executive vice president of the Children’s Defense Fund, a civil rights organization whose motto was “Leave no child behind.” Her new project, the Education Trust, became a totally new kind of progressive advocacy group, one that unapologetically used test scores—long considered suspicious on the Left—to argue that education, especially better teachers, could effectively fight poverty.
The Teacher Wars Page 21