Atlanta schools pressured to achieve higher test scores

In a room in Atlanta’s East Lake Elementary, students’ testing stats are on display like baseball players’ batting averages.

The “data room,” or “war room,” lays out district goals for the school. Staff can see at a glance how many students can fail state tests — and how many must score in the top tier — to make the numbers. Other Atlanta schools use variations of the setup.

The displays are a product of the data-driven approach pushed by Atlanta Superintendent Beverly Hall. But for teachers such as Julie Rogers-Martin, the rooms serve a second, more ominous function.

“It’s a visual pressure all the time,” she said. “It’s always in your face. It’s always a cause for concern.”

The pressure Atlanta educators such as Rogers-Martin face is rooted in a complex set of district testing goals that are harder to reach than those set by the state.

The system has rewarded school staff who were successful with nearly $17 million in bonuses since 2001. Hitting testing targets carried career and social benefits, too.

But now, the state is asking whether some schools took shortcuts to net those impressive scores.

A state probe last month identified suspicious erasures on state tests in more than two-thirds of Atlanta’s elementary and middle schools. Far more schools were flagged in Atlanta — 58 — than in any other district.

Who erased wrong answers and whether they did so to cheat is under investigation. But experts say pressure from hard-to-reach work goals and financial incentives can be a trigger for fraud.

The district-imposed testing goals are the greatest source of educators’ angst, some teachers say. In some schools, they say, making the numbers has upstaged the teaching and learning they are supposed to represent.

“You are under scrutiny every minute,” said Rogers-Martin, a fourth-grade teacher at Burgess-Peterson Elementary.

District Deputy Superintendent Kathy Augustine said the district’s data-focused approach is crucial for helping students learn. The targets aim to help educators zero in on students’ academic needs.

“Using data to drive instruction is a best practice. Data is something our parents ask for, it’s something the majority of our teachers want,” Augustine said. “I think we expect our teachers to be high performers.”

A dozen of the 18 schools that received bonuses for meeting testing targets last year showed up on the state’s “severe” list of schools with high numbers of suspicious erasures.

Augustine said that if the testing investigation finds “strong evidence” of wrongdoing, the district will take appropriate action.

All schools strive to meet state and federal standards, known as Adequate Yearly Progress. Atlanta’s goals, however, are tougher.

Each year, the superintendent approves district-wide targets based mostly on test scores. Then staff figure out how much each school must gain to move the entire system upward, said Jeff Schiller, a partner with consultants Instructional and Accountability Systems.

Schools with lower scores, which often serve the district’s poorer children, must improve more quickly than high-performing schools.

“The logic is, the further behind you are, the faster you have to learn to catch up,” he said. Targets seek to decrease the percentage of students who fail and increase the percentage scoring in the top level.

Schools that meet their targets one year face tougher ones the next. As a result, a string of steep gains and nailed targets can bring praise, but also pressure that ratchets up each year.

At one point, Hall announced a goal of having every school in the system meet 70 percent or more of its annual targets by 2008. The system has never come close.

‘Data-drowned’

Some schools hand teachers sheets with the number of children who can fail – and the number who must score at the top, “exceeds,” level – to hit district targets. Teachers submit detailed weekly reports on how each child is doing on practice tests and assignments.

The number-crunching and analyzing can be overwhelming, some say.

“We were data drowned, not data driven,” said Laurence Axtell, a 14-year veteran who left his job as a social studies teacher at Harper-Archer Middle for another district last year. “You couldn’t teach a class and do all the data and submit it on time.”

The Atlanta Journal-Constitution spoke with 10 current or former Atlanta teachers about the school environment. Several said they did not want to be named for fear of repercussions from the district.

They said some teachers are under so much stress it affects them physically. One said targets came up at every staff meeting. Another said administrators would project each teacher’s classroom scores – good or bad – on a screen for all others to see.

Some said they felt their jobs depended on test scores, even though they weren’t a formal part of teacher evaluations.

One said her principal told staff to make children pass by whatever means necessary. Two, including Axtell, said they reported to district officials hearing teachers talk about having cheated on the tests.

Asked about the complaints, Schiller said a narrow focus on numbers – and threats from school leaders – aren’t supposed to be part of the target program. Ideally, he said, a principal introduces targets at the beginning of the year and lays out a plan for improving student work. From then on, he said, talk should center on what’s happening in classrooms, not on targets.

“You don’t get more learning by focusing on targets, you get more learning by focusing on kids in those subject areas,” he said. “And good principals know that and good principals do that.”

A merit-pay pioneer

Atlanta was a pioneer in creating a system of goals and bonuses a decade ago. Schiller met Hall when she was superintendent of the struggling Newark, N.J., system and later helped her design the target approach for Atlanta, which had struggled for years with low student performance. The program was inspired by a similar effort in Charlotte, N.C.

Now, state and federal leaders are pushing similar pay-for-performance plans that link bonuses in part to scores.

As support for the district’s program, Atlanta’s Augustine pointed to a survey last year showing 83 percent of system teachers who responded said they would like incentives to be tied to their performance. Just 52 percent, however, said they felt such a link existed already.

The carrots in Atlanta’s testing program are bonuses attached to the percentage of targets a school meets. All employees in schools that hit 70 percent or more receive cash rewards of up to $2,000, depending on the targets met and their job category. Even bus drivers are eligible.

The incentive pay extends beyond schools. Hall, who has received national praise for raising student performance, earned a bonus of $78,100 last fall for the system’s success at meeting targets. Her salary and other compensation totaled $389,300 the prior fiscal year, state records show.

Some Atlanta central office staff and school reform team leaders, who are essentially area superintendents, are also eligible for raises worth thousands of dollars based partly on test scores. Student achievement counts for up to nearly a quarter of some administrators’ evaluations, which can lead to raises of as much as 7.6 percent.

When goals aren’t met, officials can receive a “needs improvement” on their evaluation. Test scores can make or break careers.

Teachers say the pressure to meet targets filters down from higher-level administrators to principals to teachers.

Succeeding on the state tests brings another benefit: status.

Each fall, a district convocation recognizes schools that have reached their targets. The event is held at a large venue, such as the Georgia Dome. The more targets schools meet, the closer their teachers sit to the front. Teachers from schools that didn’t meet targets sit farther back, off the floor.

The ritual has given rise to expressions among Atlanta teachers such as to “get on the floor” for hitting targets or to land “up in the stands” for missing them, teachers said.

One year, Peyton Forest Elementary teachers and administrators arrived at the convocation in stretch limousines — paid for by a business the district declined to name — after the school met a high number of its targets.

The school is now under investigation after the state found questionable erasures in 86 percent of the classes and subjects analyzed, making the school’s scores the third most suspicious in Georgia.

Incentive to cheat?

A 2003 study by researchers Brian Jacob and Steven Levitt found that even “relatively minor” shifts in teacher incentives affected cheating in the Chicago school district. When incentives rose, so did manipulation, the study found.

Bruce Dorris, program director for the Association of Certified Fraud Examiners and a former prosecutor, said the sort of pressure that stems from academic targets is similar to that found in the private sector.

“Wall Street and publicly-traded companies aren’t that much different,” he said. “Or a salesman needing to meet a certain quota to get a certain bonus or a raise.”

In some cases, members of professions with high levels of public respect, such as educators or ministers, have more latitude to perpetrate fraud because they don’t expect anyone to check for impropriety, Dorris said.

Atlanta officials say they see no reason to alter the bonus program.

Rogers-Martin said the Atlanta system is full of talented educators who want to help kids. But, she said, some might respond to the pressure the wrong way, by cutting corners. She said a relentless focus on often-unrealistic scores – and the piles of paperwork that come with it – are hurting teachers and students.

She plans to take a leave of absence at the end of the year.

“I really believe in this system and I believe in Dr. Hall,” she said. “But we need to correct this.”

How we got the story

The AJC spoke with 10 current or former Atlanta teachers about the climate in their elementary or middle schools. Several said they did not want to be named because they feared repercussions from the district.

The newspaper also obtained a list of the schools that met targets during the last decade, as well as five years of data on the testing targets for five schools that the state said had among the most erasures on state tests. The district provided information on details including the amount of money paid out in performance bonuses for meeting targets.

School targets

Bonuses are largely based on a school’s success at meeting testing goals.