the merit of merit pay

Big new study finds that performance bonuses for teachers boost test scores (a bit)

PHOTO: Megan Witucki
Megan Witucki, a teacher at Compass Montessori School in Wheat Ridge, works with students.

A school district leader with money for teacher bonuses faces a choice: Should she spread the money around to all teachers equally or give more to teachers who have performed best?

A new study, released by the federal government, suggests that merit-based bonuses are the way to go, as they help raise student test scores without making a significant dent in teacher morale. It offers the latest evidence that programs of this sort can help schools and students, despite the common perception that they are ineffective.

The research focuses on a federal program known as the Teacher Incentive Fund, and compares schools that gave all teachers an automatic 1 percent bonus to those schools that gave bonuses based on classroom observations and student test scores. About 130 schools across 10 districts were randomly assigned to one of the groups.

The result? The schools that gave performance bonuses boosted student test scores throughout the four years of the study, between 2011 and 2015.

The differences between the two groups of schools weren’t big, and they were only sometimes statistically significant. For instance, in year three of the study, performance pay increased student test score performance by 2 percentile points.

A graph showing the share of teachers getting different size bonuses in districts that implemented performance pay.

But because the program was relatively cheap to implement, performance bonuses were very cost effective. They offered a better bang for the buck than class-size reductions, the researchers estimated.

And what about fears that performance pay would dampen teacher morale or reduce collaboration as teachers competed with each other for a fixed pool of bonus money? There was little evidence of that. According to teacher surveys, the program had only a limited impact on their job satisfaction, interactions with colleagues, or school morale. In the initial year of performance pay, there seemed to be small dips in morale and collaboration, but in year three the effects were actually positive.

The program also led to small increases in teacher retention, which may help explain the positive findings.

The study comes with an important caveat: It relies on test scores alone as a measure of learning. If the incentives pushed teachers to raise scores by less desirable means, like cheating or “teaching to the test,” that wouldn’t say much about the success of the policy.

The research adds to the hotly contested debate about how teachers should be paid. High-profile research in New York City and Tennessee from several years ago found disappointing results for merit pay, solidifying a conventional wisdom that the approach doesn’t work for teachers. Other districts, like Denver, have struggled with the design of a performance pay program and have seen inconsistent results.

But some recent studies have re-opened this debate. An overview of research last year showed that performance pay leads to small boosts in test scores, and a number of studies have found that added pay can help keep more effective teachers in the classroom.

It’s not entirely clear why the studies have reached such different conclusions. But performance-pay programs that judge teachers by test scores alone and only measure short-term effects tend to be disappointing. Programs like the Teacher Incentive Fund and those in Minnesota and Washington, DC, which use performance pay as part of a broader system for evaluating teachers, often produce more positive results.

That doesn’t mean that districts will heed the latest research. In fact, even in the new federal study — which found benefits for students — less than half the districts said they planned to continue offering performance bonuses after the Department of Education grant ran out.

Measure of Success

State ratings identify 163 Colorado schools in need of improvement

PHOTO: AAron Ontiveroz/The Denver Post
Students in kindergarten on the first day of school at McGlone Academy.

More than 160 Colorado schools received one of the state’s two lowest ratings, making them eligible for additional assistance but also vulnerable to intervention if they don’t improve student performance.

The watch list comprises 9 percent of Colorado’s 1,800 schools and educate roughly 74,000 students, or 8.5 percent or the state’s almost 900,000 students. That means the vast majority of students in the state attend a school with one of the two higher rankings on the four-point scale.

The State Board of Education finalized the ratings Wednesday. The state gives separate district-wide ratings, which were finalized last month.

“The state’s accountability system is built on the premise that all students should receive a high quality education and graduate ready for college or careers,” Katy Anthes, Colorado’s education commissioner, said in a statement. “Our goal is to give all students a chance to excel. These designations allow us to identify struggling schools that may need more support to help students achieve their highest aspirations. And they also highlight successful schools so that other schools can learn from them.”

All public schools receive a state rating, known as the School Performance Framework report, each year. It’s based largely on student scores on the state’s English and math tests. Student growth, or how much students learn year-to-year compared to peers with similar results on state tests, carries most weight. High school graduation and dropout rates are also factored in.

Colorado Department of Education

There are four ratings: performance (the highest), improvement, priority improvement and turnaround (the lowest).

Schools and districts that have one of the lower two ratings are placed on a watch list and have five years to improve before facing state intervention. Schools on the list are eligible for grants for leadership training and help from outside consultants, but if change doesn’t come fast enough, the state could hand over control to an external manager, require conversion to a charter, or close schools.

Earlier this fall, the State Board of Education ordered the Adams 14 school district, based in Commerce City, and two schools in Pueblo in southern Colorado to turn over control to external managers after earlier intervention efforts did not produce enough improvement.

Colorado is still figuring out what effective intervention looks like and if outsiders can make a difference for students that existing leadership has not been able to achieve.

Most Colorado schools maintained the same rating they had in 2017, with 15 percent moving down at least one level and 14 percent moving up at least one level. Eighteen schools improved enough to get off the state watch list, which is often known as the “accountability clock,” some after initial state intervention last year.

Six schools are entering their eighth year on the watch list: Aurora Central High School, Adams City High School, Aguilar Junior-Senior High School in the tiny Aguilar district in southern Colorado, Hope Online Learning Academy Elementary School in Douglas County, Heroes Middle School, and Risley International Academy of Innovation, the last two both in Pueblo.

Two are entering year six: Central Elementary School in the Adams 14 district and Minnequa Elementary School in Pueblo.

Another four are entering year five, now the last year to improve before state intervention: Manual High School and Montbello Career and Technical High School in Denver, Mesa Elementary in the Montezuma-Cortez district in southwest Colorado, and EDCSD: Colorado Cyber School in Douglas County.

In the past, some schools received more time to improve because the “clock” was paused for several years as the state changed assessments. But now there are no more extensions beyond year five.

Of the state’s 42 online schools, a little more than half received one of the top two ratings, and 31 percent did not report enough data for the state to grant a rating. Colorado has more stringent regulations of online schools than many states, but there is an ongoing debate about how well these schools serve students.

About 84 percent of the state’s 247 charter schools received one of the top two ratings, compared to 89 percent of all Colorado schools. Twenty-six charter schools, or 10.5 percent, received one of the lowest two ratings.

Look up your school here:

Literacy tutors needed

Detroit enlists volunteer tutors before third-grade reading law takes effect

PHOTO: Anthony Lanzilote

Detroit’s school district is asking the community for help getting students reading at grade level. The superintendent is hoping volunteer literacy tutors will prevent a critical mass of third-graders from being held back under the state’s tough new reading law.

“We need your help,” Superintendent Nikolai Vitti said, making an appeal for volunteers during a school board meeting Tuesday night. “Our teachers and our principals and our schools alone will not be able to ensure that every student is at third-grade level without your help.”

Which is why the district is working with two community advocacy groups, Keep the Vote/No Takeover and the National Action Network, to launch the Let’s Read program, geared to K-3 students. The program is slated to begin in February — less than a year before the reading law takes effect. Once it does, during the 2019-2020 school year, Michigan third-graders who aren’t reading at grade level will be held back.

In the Detroit district, where proficiency levels on state exams are extremely low, the consequences could be dire. During a community forum last week, Vitti said that the law could hold back as many as 90% of Detroit third-graders, though Michigan’s education department has yet to define what it means for a student to be reading at grade level. At the forum, though, he noted exemptions from the law for such as students with special education needs and those who speak little to no English.

The Let’s Read volunteers will be assigned to individual students based on need. They will read with the children and help them with book selections.

Helen Moore, a longtime community activist who represents the two community organizations behind the volunteer effort, urged people to sign up during the public comment period of the meeting.

“I know our students will succeed, because they’re brilliant,” Moore said. But they and their parents need help, she said.

Vitti said the volunteer cohort is one of many literacy-building efforts underway. In addition, he said that every district school will hold family literacy nights and that its Parent Academy will expand its classes that teach parents how to help their children with reading. A community-wide event to teach Detroiters about the reading law — and what they can do to help — will also be held.

Moore said the word is starting to get out about the Let’s Read program, noting: “The telephone has been ringing like crazy. And now the suburban districts want to be part of it.”

The focus, though, is on Detroit, she said.

Want to volunteer: You can fill out a form here, or call 313-873-7884.