The Teacher Wars (31 page)

It was almost lunchtime, and the kids were yawning and fidgety. Arpino stopped every few minutes to enforce rules on how to sit. “I'm waiting for all my scholars to sit in criss-cross-applesauce!” she demanded. “With their hands folded and back straight. We have to grow our brains for first grade! Because why? Why do we grow our brains?”

“For second grade!” said Melvin.

“Yeah, but what's the
big
goal?” she asked.

“College!” cried a little girl named Chanel.

“Yes, college,” Arpino repeated. “And then we can change the world.”

Later on I visited the fourth-grade math class of Tarik Walmsley, a lanky University of Washington graduate who was homeschooled at his students' age. Walmsley's lesson was on the idea that multiplication and division are inverse operations: that 8 × 2 = 16 and 16 ÷ 2 = 8. He passed out small plastic blocks and had the kids arrange them in various groupings: four groups of four blocks each, two groups of eight blocks each. Student behavior had been a challenge,
Walmsley told me. One girl sometimes got up from her seat to dance across the classroom. A boy with a special-ed diagnosis could answer problems on paper but had trouble speaking up in front of his classmates. On a quiz, he wrote Walmsley a note: “Teacher, you think I'm stupid, but I'm not.”

On the wall was a chart showing a ladder, each level representing one behavioral demerit. Step 1 is a warning. At Step 3, a child is sent to the “icebox,” an isolated chair at the back of the classroom. By Step 5, a parent is notified, and the child is removed from the classroom. Each student's name was written on a wooden clothespin, and as he or she accrued demerits, the pin moved up the ladder. Like Arpino with her kindergarteners, Walmsley spent an extraordinary amount of time policing how his fourth graders sat. Were their eyes “tracking” the teacher? Were pencils resting in the pencil groove of the desk? He didn't hesitate to give demerits for small infractions. “Remember how I was talking about chocolate milk? How milk and chocolate are our
products
?” he asked the students, referencing the previous day's multiplication lesson. When a boy named Anthony answered, “Yes!” he earned a demerit for speaking out of turn. By the end of the period, Anthony's clothespin had moved up the ladder, and Anthony was sitting in the icebox, scowling.

After class Walmsley said, “Being stern is not what I grew up with. But it feels useful and fair.” He acknowledged he was still learning how to teach. His TFA mentor had said he was performing like only a “typical” teacher; his students were not yet demonstrating the “dramatic” or “path-breaking” levels of achievement TFA expects, and which the program literature told recruits would eventually close the national achievement gap between affluent and poor children.

While TFA was once criticized for producing teachers with little idea of how to manage a classroom or create a lesson plan, today TFA offers its corps members a prescriptive set of directions on how to teach, and even how to
think
about teaching. It is called Teaching as Leadership. Teach for America is often lauded as an alternative to hidebound graduate schools of education, but the central idea of Teaching as Leadership is borrowed from two theorists,
Jay McTighe and Grant Wiggins, whose ideas are popular at some traditional teachers colleges, too. McTighe and Wiggins call their concept “backward design,” and TFA calls it “backward planning.” According to the
Teaching as Leadership
sourcebook, the first step in backward planning, whether one is teaching public school or running a company, is for the visionary leader to come up with “a big measurable goal.” For a CEO, that goal might be to sell one million gadgets. For teachers, TFA recommends goals like “All my first graders will advance two reading levels within one school year,” or “I will put my sixth-grade students on a path to get accepted into competitive high schools in our city.” Citing “Pygmalion in the Classroom,” the 1968 study showing that children earn higher test scores when their teachers have high expectations, the
Teaching as Leadership
book tells recruits not to worry if the goal seems too ambitious—even “crazy”—for a classroom full of students who are far behind grade level. Thinking big is the point. And it warns teachers against perhaps worthwhile but unquantifiable goals like “I will turn my English students into lifelong readers” or “I will develop my history students' sense of citizenship.”

The next step is to determine what data can be collected to prove the big goal has been achieved. The data will almost always be test scores, from either a state standardized test, a district test, or a test the teacher finds or creates on her own. A sixth-grade teacher might know that students need at least an 85 on the state's end-of-year English test to be considered for competitive high schools. So she will pore over exam questions from previous years and will target every lesson plan, homework assignment, and student assessment toward building the skills that will enable her students to do well on that test. If she has to, she will host evening tutoring sessions at McDonald's, tempting her students with free food. She will write a weekly class newsletter with celebratory “shout-outs” to students who perform well on quizzes. When she calls or visits a student's home, she will seek to “invest” parents in the big goal.

During the recruitment and selection process, TFA seeks corps members who are likely to embrace this backward-planning, data-driven mind-set. The organization constantly tracks which recruits produce the largest test score gains for their students, reviews those
teachers' characteristics, then looks for new candidates who display similar achievements and behaviors during the interview process.

Teach for America was founded during an era of teacher shortages and promoted on the basis that it was filling a great need. Today, with teacher layoffs and high unemployment, the organization cannot justify itself on the same grounds and instead explicitly advertises its corps members as more effective than veteran teachers.
The research consensus on TFA suggests that corps members are about equally effective at raising students' test scores as teachers from all other pathways, though better in math than in reading and writing. A September 2013 study from Mathematica Policy Research found that TFA middle and high school math teachers outperform other math teachers in their schools, though only by the equivalent of students gaining 3 points on a 100-point test. The researchers could not discern exactly why. Like other research on teacher credentials, the study found that regardless of the pathway into the classroom, teachers who majored in math or who had attended selective colleges did not seem to significantly outperform other teachers with less impressive résumés.
^*2
So if the “best and brightest” theory isn't true—if traditional meritocratic credentials aren't the reason that TFA teachers are good at their jobs—then what accounts for many corps members' success?

The work of John Hattie provides some clues. He is an education researcher from New Zealand who has reviewed eight hundred meta-analyses that summarize the results of over fifty thousand education studies. Hattie has found that completely separate from a teacher's demographic traits, a few specific teacher behaviors—including some emphasized by TFA—powerfully influence student achievement. A wealth of research indicates that one of the best
things a teacher can do for her students is to set high, individualized expectations for each one of them, regardless of a child's past performance or whether he comes to class with a label such as low-income, special education, or learning disabled. Effective teachers believe all children can learn—a fundament of the TFA philosophy too—and reject the idea of intelligence as an inborn trait, instead seeing it as something a teacher can develop in every student. In general, academically ineffective teachers are those who set the bar too low; some evidence suggests that half of what is taught in most classes is already known by most students. That brings us to another teacher behavior Hattie identifies as potentially transformative, and which TFA promotes: formative assessment. To avoid teaching children what they have previously learned, teachers should assess students at the beginning of the school year and at the beginning of new units, to identify their strengths and weaknesses. Students should be quizzed again when units end, to determine if concepts and skills have been successfully taught.
A cache of studies from cognitive scientists confirms that students score higher on end-of-course standardized tests when they have been periodically quizzed along the way.

Although education research seems to confirm some of TFA's practices and mind-sets, it calls others into question, particularly those having to do with student discipline. When teachers provide constant, controlling behavioral feedback, as Arpino and Walmsley were being taught to do, they waste precious time they could be spending giving feedback related to the academic content of the lesson, which is far more powerful in terms of raising student achievement. One of the challenges of training a new teacher, Hattie writes, is convincing her that “
developing a strong desire to control student behavior can be inconsistent with implementing many conceptual approaches to teaching.”

There has been little convincing research done on those “no excuses” teaching strategies: incentive systems (pizza for good behavior and high test scores); the focus on children's posture and eye contact as teachers read or lecture; and the school uniforms and silent hallways. Yet this type of teaching has exploded in prominence since the mid-1990s, driven in large part by the strategies and
rhetoric used at the KIPP charter schools and adopted by TFA and many of the other charters at which its corps members and alumni increasingly work.
^*3

The KIPP schools (pronounced “kip”) are the most celebrated in America. They were founded in 1994 by two TFA alums, Dave Levin and Mike Feinberg, who, as struggling first-year teachers in Houston, became entranced by the classroom strategies of Harriet Ball. A magnetic, six-foot-one African American woman, Ball seemed to work miracles with her African American fourth graders. Her students exuberantly sang songs—mnemonics about what they were learning, as well as exhortations of
why
they were learning (
'cause knowledge is power, and power is money, and I want it!
)—but when Ball snapped her fingers, they went dead silent. If children acted up or didn't do their homework, she threatened to transfer them to another teacher's classroom. They were lucky, she told them, to have her as a teacher, and they better not waste the opportunity.

Ball emphasized education as a privilege and literacy as the pathway to personal financial empowerment—an ideology with a long, proud history in the African American community, traceable to Booker T. Washington. Her high expectations for her students echoed the promise of Rhody McCoy, the superintendent in Ocean Hill—Brownsville, that children would learn if teachers set the right tone. In the 1990s a multiracial group of Generation X education reformers began to adopt and translate these strategies. Levin and Feinberg named their schools KIPP—the Knowledge Is Power Program—after the refrain in Ball's song. Her “no excuses” methods supposedly proved, as Wendy Kopp has written, that “
education can trump poverty,” as long as a teacher accepts her responsibility as the “key variable” driving student outcomes. “We”—not parents,
not neighborhoods, not school funding or health care or racism or stable
housing—“control our students' success and failure,” states
Teaching as Leadership
.

This ideal of the all-powerful individual teacher, solely responsible for raising student achievement in measurable ways, soon transcended start-ups like Teach for America and KIPP to become the foundation of national education policy making during the Obama years. It received a big boost from a new way of evaluating teachers and schools, called value-added measurement.

On May 20, 2003, Kati Haycock of the Education Trust appeared in front of Congress to share her opinion on how No Child Left Behind could be improved. Haycock had been instrumental in pushing for NCLB. Now she had a new message for lawmakers: It was clear that individual teachers, even more than standards or schools themselves, were “the number one ingredient of high achievement” for kids. She cited a body of research from a University of Tennessee statistician named William Sanders. Using a technique called value-added measurement, Haycock said, Sanders had proved that a child with a sequence of good teachers could demonstrate up to 50 points of gain on a 100-point standardized test, “
the difference between entry into a selective college and a lifetime working at McDonald's.” Teachers should be evaluated, she said, on whether they “produce student learning gains.”

Sanders's claims were stunning. The most important education research of the 1950s and 1960s had been conducted not by testing experts, but by psychologists and sociologists. Kenneth Clark and James Coleman had looked at a broad range of factors that influenced children's school performance and overall well-being: how many books their parents owned, what toys they played with, whether schools had science laboratories or libraries.
When the older generation of researchers tried to pinpoint what made a teacher successful, they often looked for particular personality traits like warmth, extroversion, and conscientiousness. But the explosion of state testing programs since the 1970s provided researchers like Sanders with an unprecedented data trove. Statisticians and economists
have used this achievement data to ask a much narrower question: Which teachers raise or lower a child's test scores?

Value-added measurement is the method researchers developed to find an answer. In its relatively crude, early form, value-added simply used a student's score on an end-of-year standardized test to predict her score on the following year's exam. Teachers who presided over larger-than-expected jumps in scores earned above-average value-added ratings. (For example: Sarah scored an 89 in third-grade math. The typical child who scores 89 gets a 91 next year, but Sarah scored a 93 in fourth grade. Those 2 points of unpredicted achievement gain are attributed to her fourth-grade teacher and are computed into the teacher's value-added rating.)