Tennessee’s overhaul of teacher evaluation in 2011 led to sharp backlash and higher teacher turnover in the immediate aftermath, but a new study in the American Educational Research Journal finds that turnover may have helped low-performing urban schools in the long run.
That’s because highly effective teachers left their schools at 15 percent lower rates than did their least-effective peers, according to a team of researchers led by Luis Rodriguez of New York University.
Tennessee’s teacher rating system, among the first developed under the federal Obama-era Race to the Top competition, drew criticism and even a lawsuit because it tied teachers’ evaluations in part to their students test-score growth. But it became a model for those of other states like Delaware, North Carolina, and Connecticut. Echoes of the system have even shown up in the Democratic presidential debates. But district leaders are still trying to draw lessons in how best to use multi-part, value-added teacher-evaluation systems like Tennessee’s to improve instruction without driving away good teachers.
Rodriguez analyzed state teacher evaluation and turnover data before and after the state implemented its Tennessee Teacher Acceleration Model, or TEAM, as a default evaluation system along with several alternate evaluation models, made as part of the Volunteeer state’s bid for federal Race to the Top grants.
TEAM incorporated classroom observations, portfolios, and student surveys, as well as student test-score growth and achievement data in reading and math into teachers’ effectiveness ratings in tested grades, changed tenure to make it easier to fire ineffective teachers, and also piloted retention bonuses to keep highly effective teachers in high-need schools.
The researchers found that before the new evaluation system was put into place, struggling schools faced higher turnover among their highest-performing teachers (based on standard deviation above the average teacher rating), while their least effective teachers were likely to stay, as the charts below show in both English/language arts and math. By contrast, the least-effective teachers were most likely to leave high-performing schools, likely through being transferred or counseled out.
After the teacher evaluation system changed, there was little change in the turnover patterns for higher-performing schools—but in struggling schools, the turnover patterns totally flipped:
The least-effective teachers in both reading and math (those whose ratings had been a full standard deviation below the average) had the highest probability of leaving low-performing schools. Moreover,the turnover rates for the most-effective teachers in low-performing schools were about the same or lower than those of their colleagues in higher-performing schools in math and English/language arts, the researchers found.
In an associated study, Rodriguez and his associates also probed the state’s retention bonus program, which was piloted in 2011-12, the first year of the new evaluation system. Under the pilot, if a teacher working in a low-performing “priority” school had been rated as highly effective in previous years, he or she could qualify for a bonus for remaining at the school another year.
“What we found was that the retention bonus did result in increased retention for these high-performing teachers who were working in low-performing schools,” Rodriguez said. Highly effective teachers in schools that used retention bonuses had turnover rates 20 percent lower than their least-effective peers, suggesting there was a benefit above and beyond the improvements from the new evaluation system overall.
“However, some of the patterns that we were seeing in terms of turnover and retention among high-performing teachers lasted way beyond that point—in fact, most of the differences we’re seeing were concentrated in 2015—so it seems like there’s some other mechanisms going on that weren’t wholly tied to the retention-bonus program,” he said. “It seems like the patterns that we’re seeing across across both the evaluation system overall and the retention-bonus program were running in the same direction.”
On average statewide, highly effective teachers had 15 percent lower turnover rates than their least-effective colleagues as a result of the new evaluation system, but these effects were concentrated in low-performing and urban schools. Rodriguez noted that in these areas, district leaders may simply not have a large enough labor market to replace teachers who are not effective instructors, or it may cost significantly more to bring in new teachers than to improve the practice of existing educators.
Nearly a decade later, 3 out of 4 Tennessee teachers now believe Tennessee’s teacher ratings overhaul has improved their instructional practice. And in a separate study of the TEAM system, Brown University researchers found the state’s efforts to highlight highly effective teachers as models and coaches for other teachers helped both rural and urban educators improve their instruction.
“You know, it’s great that we’re seeing relative retention of more-effective teachers once the evaluation system is in place. That’s great. But ultimately we can’t rely purely on the evaluation system,” Rodriguez said. “We have to provide other mechanisms and other policies that will sustain the highly effective teaching workforce: things like differentiated pay and retention bonuses, yes, but also coaching and mentoring and allowing alternative entry to the workforce. Because if you wholly rely on evaluation for accountability and dismissal of low-rated teachers, you really can’t fire your way into a better teaching workforce as a good long-term strategy.”
Chart Source: AERJ
A version of this news article first appeared in the Inside School Research blog.