Correction: An earlier version of this article incorrectly said that 35 percent of the teachers rated “highly effective” last year were in Ward 3, the city’s wealthiest, and that 5 percent were in Ward 8, the city’s poorest. In fact, 35 percent of the teachers in Ward 3 and 5 percent of the teachers in Ward 8 were rated highly effective. This version has been corrected.

They met on an icy afternoon, Clay Harris, an elementary math teacher at the end of a hectic day, and Eric Bethel, one of the city's new master educators, there to render a verdict on Harris’s teaching that could determine whether he keeps his job.

In polite, awkward silence, they walked to Harris’s empty classroom at Beers Elementary School in Southeast Washington and settled in kid-size chairs at a low, yellow table.

Bethel set up his laptop. Harris took out a piece of paper for notes and began tapping his pencil on it.

“I didn’t do everything perfectly,” he said almost apologetically.

Bethel smiled. “No one does,” he said.

This conference, one of hundreds underway across the District, offers a rare glimpse into the hypersensitive center of an education reform movement that has taken aim squarely at teachers.

The idea, aggressively embraced by the Obama administration, is as straightforward as it is controversial: Teachers are the main factor in student growth — more than poverty, parents, curriculum, principals or other circumstance. Improve the quality of instruction, the logic goes, and you will improve the public schools, a conviction that has led districts to adopt more-rigorous ways of evaluating teachers.

The District’s system — called IMPACT, now in its second year — is becoming a national model, even as unions and some experts question the wisdom of staking careers on it. Last year, then-Schools Chancellor Michelle A. Rhee fired 75 teachers who received poor evaluations. Firing a teacher was a rarity under the old system, which tended to rate teachers highly even as the city’s schools accrued one of the nation’s worst academic records.

Under the new system, the annual rating for teachers in certain grades depends heavily on how much their students’ test scores improve, a metric that many experts consider unreliable.

But the heart of the new system is a set of nine standards — from explaining content clearly to time management — that are supposed to represent years of research on the elements of great instruction. During five half-hour classroom observations throughout the year, most unannounced, principals or master educators assess how well teachers incorporate the standards into their teaching

After each observation, there is a conference. It’s the moment when the reform agenda meets the teachers expected to carry it out, when teachers so often praised in emotional or even artistic terms — for effort, devotion, creativity— confront a new, more technical set of measures.

“There is this ‘Bless your heart’ problem in the teaching profession,” said Jason Kamras, the key architect of IMPACT. “It’s, ‘This is so hard, so bless your heart for trying.’ That’s not how you become a real profession. We need to be honest about that conversation.”

Mr. Bethel and Mr. Harris

And so Mr. Bethel and Mr. Harris, as they called each other, settled down to try.

Bethel, crisp in gray slacks and shiny loafers, introduced himself. A former D.C. elementary math teacher with a master’s degree, he told Harris that he understood the anxiety about the new system, because he went through it as a teacher last year.

Though initially skeptical, he found the new system a huge improvement. Instead of generalized praise, he received concrete advice based on a clear set of principles — a specific language for discussing the craft. Things he had done instinctively, such as checking whether students understood the lesson, were now expressed in the standards, which encouraged him to practice them more systematically. His teaching got better.

Still, Bethel had wavered about the evaluator job, wondering if he had the stomach to deliver difficult critiques to more-experienced teachers. He finally decided he could perhaps “bridge the gap” between teachers’ anxieties and a tool he was certain could make even a great teacher better.

In Harris, he had a fine test. A part-time cartoonist of artistic temperament, Harris described IMPACT as “a sword hanging over my head.” He majored in math in college, and he has been a D.C. teacher since 1989, since December at Beers. But he worried that his freewheeling teaching style would bomb under the new system. And in his case, classroom observations would count for 75 percent of his overall rating.

He tapped the pencil.

“So,” said Bethel, who had observed Harris teaching a fourth-grade class. “I can tell you the key things you have to celebrate or key areas to grow on.”

“You can start off by what I did right,” Harris said. “Just start there.”

Harris scored well in creating a “supportive, learning-focused classroom community.”

While many teachers view the standards as a kind of “to do” list — one teacher, for instance, thought she had to lecture her students on being supportive — Harris had demonstrated the principle, Bethel said.

He had observed an easy rapport between Harris and his students that was evident in their attentiveness to Harris’s instructions.

“What kind of solidified it was that you had a student, she wanted to get up, and you let her move around some but were able to keep a good balance,” Bethel said. “That was great.”

Harris nodded and took notes.

Beers Elementary, a 350-student school off shaded Alabama Avenue, is neither one of the best nor one of the worst schools in the city. Last year, 44 percent of its students scored proficient in math, 46 percent in reading. Most of its students receive free or reduced-price lunches, a standard measure of poverty.

According to Nathan Saunders, president of the Washington Teachers’ Union, poverty is among the reasons that IMPACT is unfair: It is more difficult for teachers in such schools to earn a decent rating than for those in wealthier areas.

Last year, about 35 percent of the teachers in Ward 3, the city’s wealthiest, and five percent of the teachers in Ward 8, the poorest, were rated “highly effective.” Beers is in Ward 7, also one of the poorest.

Like most teachers he knows, Harris tries to compensate for his students’ challenges. He serves as a mentor and tutors on weekends. But under the new system, such extracurricular gestures barely register.

The disagreement

Instead, Bethel dissected Harris’s teaching as a doctor might be scrutinized for technical skill.

Bethel’s most serious concern involved how Harris had taught his lesson on the commutative property, the math law that says 3 + 5 is the same as 5 + 3.

Bethel described what he had observed: Harris had written on the whiteboard: 855 + 319 = 1,174. Underneath, he had written four problems, such as 855+300+19+1 and 800+50+5+150+150+19.

Students were supposed to work the four problems and discover the underlying math law. But had the students done that, Bethel said, they would have discovered a different concept.

“So basically you showed them decomposition,” Bethel said. “That was the discovery, not so much that order doesn’t matter,” which was the objective.

Harris sat up. He raised his eyebrows, and in slightly exasperated tones, began offering his critique of the critique.

The problems on the board, Harris said, were just a warm-up to review the concept of place value. But it soon became clear that the students were struggling simply to add. And in that moment, Harris said, he decided to scrap the objective and rehash place value.

“It seems like I’m getting penalized possibly because I didn’t do that exact lesson I set out to do,” Harris said, explaining that many of his students were three grade levels behind. “I’m trying to get the kids up to a speed where they could learn that lesson. A lot of our kids, they fundamentally don’t know.”

He apologized for being politically incorrect. Then he continued. He said that all children do not learn equally and that “our children, especially, don’t learn equally.”

Bethel nodded. “I hear what you’re saying,” he said.

But the truth was he did not agree. From what Bethel had observed, the kids were simply confused, not unable to add.

Though it was not easy, Bethel said, it was possible to teach the objective while working on students’ weaknesses. Delicately, smiling, he offered a suggestion.

Harris might have divided the students into three groups, giving each one the same simple addition problem, but with the order of the numbers rearranged. Each group would solve their problem. Then they would share their answers.

“Then you give the students a chance to say, ‘Hey, we all have same answer!’” Bethel said, lighting up. “Then you let them discover that the numbers were arranged in different ways. Then you’re getting at the key concept, that order doesn’t matter.”

It was an elegant solution, Harris acknowledged.

They moved on to the last of the nine standards: engaging students in rigorous work. Harris read his score.

“It says ineffective,” he said, incredulous.

Bethel explained that the warm-up problems, while rigorous, were aimed at the wrong objective. They went over it again. Harris sighed.

“I’m sorry,” Bethel said. “I hate that this is discouraging. I really do. You’ve had good ideas, really.”

Harris interrupted.

“What I’m trying to convey to you, Mr. — ” he paused — “Eric — is that most lesson plans, the best ones, no matter how pinpoint-precise I plan it, the lesson will deviate. It will deviate because there is always some other rock I have to overturn to look at.”

Bethel gave him the final score, which was low. If the trend continued, Harris realized, he could lose his job.

“It’s just — I don’t feel that I’m putting in ‘minimally effective’ effort at all,” he said.

For Bethel, this was most excruciating part of the job. He began shutting off his computer.

“This does not measure your effort,” he said, packing his bag. “But I do see your effort, Mr. Harris.”

“So — what is this measuring?” Harris asked.

“It’s measuring the effectiveness of that effort,” Bethel said. “This is not a reflection of your passion for education, your love for students. Not at all.”

Which for Harris was precisely the problem and for Bethel was part of a difficult, painful solution.

As he left, Bethel offered to help Harris with lesson planning, a gesture that would not count on Bethel’s own evaluation. Harris leaned back in the little chair. He pursed his lips.

“I don’t think you’re being personally unfair, it’s just — ” he paused. “I’m going to look over it again. I know where I could improve. So. Yeah. It was nice talking to you.”