Offering a one-time bonus could help keep high-performing teachers in high-needs schools and raise student scores, according to a new study published in Economics of Education Review.
Researchers from the University of Georgia, New York University, and the University of North Carolina, Chapel Hill looked at Tennessee, which put this kind of bonus program in place in 2012 for teachers working in the state’s Priority Schools—those with student test scores in the lowest 5 percent of the state.
Teachers in those schools who scored at the highest level on the state’s effectiveness ranking were offered a one-time, $5,000 bonus (a 10 percent salary increase, on average) if they committed to returning to the same school the next year.
The program produced results, according to the study.
Previous research by Matthew G. Springer, a professor at the University of North Carolina, Chapel Hill’s school of education, and one of this study’s authors, found that the extra money did keep more high-performing teachers in these high-needs schools. This current study found that student performance improved as well: Reading scores at these schools increased by about .1 standard deviation, or between 8 and 11.5 additional weeks of learning, compared to similar schools that didn’t have the bonus program. Math performance improved as well, though the difference was only marginally statistically significant.
These results add to the contentious body of research on performance pay. Past studies have found mixed results, though a study on federal performance pay from 2017 found that it led to an increase in students’ math and reading scores. States and districts have debated these policies for more than a decade, but they remain controversial among teachers and some education leaders.
Teasing Out Types of Performance Pay
Not all performance pay systems are alike, though, said Springer. “It’s important that we tease out the different types of incentive pay systems that there are, because some people will just say, ‘Merit pay doesn’t work, or merit pay works.’ In reality, we need a much more nuanced discussion of it,” he said.
A retention bonus—the tool used in this study—works differently than a performance incentive that promises to reward teachers who raise their students’ scores, said Walker A. Swain, an assistant professor of education policy at the University of Georgia, and the study’s lead author.
Retention bonuses are aimed at affecting the composition of a school’s teaching staff, asking an already effective teacher to make the choice to stay when she might otherwise move to a different school. By contrast, said Swain, a performance incentive that rewards teachers whose students’ scores improve is asking a teacher to change her practice—something that may require more time or effort to achieve.
With a retention bonus, he said, “we’re not asking you to up and transform any of the practices that you’re doing. We’re just saying, ‘Don’t leave.’”
The effects have the potential to reach beyond higher test scores, said Swain. “At the core of the policy idea here is that you’re able to improve the stability of the school and keep potential instructional leaders in the school, people who have positive peer effects on other teachers,” he said. And in many cases, said Swain, offering bonuses could work out to cost less than the recruitment and induction associated with teacher turnover.
Spending more money on teachers in low-performing schools is an equity issue, he said. In Tennessee, low-scoring schools tended to serve students from low-income families and students of color. “If you’re losing your effective, experienced teachers and replacing them with new teachers who get paid at a lower rate ... then the amount of money that’s getting spent on these kids in high-poverty schools is less than the kids who are getting taught by the $60,000 teacher with a master’s and performance pay,” he said.
There is a caution to this kind of performance pay, though. If teacher effectiveness ratings are based mostly or solely on student test scores, teachers of high-performing students will receive bonuses—but it won’t be clear whether the teacher was necessarily the cause of higher student achievement.
“There’s definitely noise in test score-based measures,” said Springer. But in Tennessee’s case, he said, the composite measure of teacher effectiveness makes the rating “a little more stable.” In tested subjects, achievement data makes up half of the teacher evaluation score, while the remaining 50 percent is made up of observations, conferences, and reviews of prior work.
A version of this news article first appeared in the Teaching Now blog.