Guest post by John Thompson.
In the pseudo-documentary/propaganda film Waiting for Superman, we were told of something called “the dance of the lemons,” in which ineffective teachers are shuffled from one school to another rather than fired. One of the big ideas the reformers have dreamed about is the creation of a sort of “dance of the lemons” in reverse, whereby, instead of the worst teachers moving around leaving a trail of poorly taught students, we somehow lure the best teachers to go where they are most needed. After all, if teacher quality is indeed the largest factor within our control, we must somehow be able to direct better teachers to places where they are needed most. The results of the latest study ought to give even the most enthusiastic social engineer a bit of a reality check.
The latest study by the Institute of Education Sciences, “Access to Effective Teaching for Disadvantaged Students,” by researchers from Mathematica and the American Institutes for Research (AIR) documents once again a terrible disparity between effective teaching in low and high poverty schools. It showed that “Teachers of FRL (Free and Reduced Lunch) students have lower value added than teachers of non-FRL students on average, with statistically significant differences of 0.034 standard deviations of student test scores in ELA and 0.024 standard deviations in math.” The AIR estimates that this would translate into a two percentile difference in student performance.
Of course, it is impossible to estimate how much of that disparity is due to lower quality teachers in low-income schools and how much of it is due to the conditions that undermine effective instruction in poor schools. Common sense says that the inequitable results are due to both, lower quality teaching that results from the awful conditions that drive so many of the best teachers out of poor schools, and the peer effects and misbehavior in those schools. Especially in inner city middle schools, the single most likely culprit is the chronic disorder, due to traumatized children acting out their pain, which makes it much harder for good teachers to teach effectively.
It was worth investigating whether the quickest and easiest route to school improvement, monetary incentives, made any sense. In a rational world, financial incentives might persuade enough of the best teachers to transfer to the tough schools and not leave after collecting their bonuses. If they increased student performance somewhat, the carrots would be worthwhile.
In an age of “reform,” however, it is not so simple. Teachers renowned for adding “value” in lower-poverty schools would have to produce increases poor students’ performance or risk termination. In an era of value-added evaluations, it seems inevitable that some teachers who volunteer for the more difficult task of raising test scores in the toughest schools will find their careers damaged or destroyed, through no fault of their own, because they sought to help students with lower test scores.
In other words, the idea that financial incentives can improve teacher quality in an age of high-stakes testing is dead on arrival unless top teachers can produce gains in the tough schools that are comparable to gains in lower-poverty schools. In theory, that might happen in elementary schools. In inner city middle and high schools, that ain’t gonna happen.
Mathematica, working with the TNTP, studied whether a large financial incentive (up to $20,000 paid out over two years) known as the Teacher Transfer Initiative (TTI) could encourage high-performing teachers to transfer to low-performing schools, whether they would be successful in raising the achievement of their new students, and whether the top teachers would remain after the payments end. “Transfer Incentives for High-Performing Teachers,” by Steven Glazerman et.al, also asked whether the top new teachers would become leaders and otherwise help improve their new school.
The process of setting up the study should have convinced anybody, even the true believers at the TNTP, that transfer incentives (even if they work) cannot be more than a niche solution in an age of accountability-driven reform. They investigated 59 districts and found ten large school systems where the TTI was logistically possible. They invited the top-performing teachers in 695 higher-performing schools and offered them incentives. Teachers in more than 90% of those schools turned them down. Only 81 teachers with high value-added ratings were transferred to 64 schools (out of 205 schools that had been identified as low-performing.) They also identified 30 high-performing teachers who stayed in their challenging environments and who were made eligible for performance incentives.
A footnote reported that the ratio of high-performing teachers to the number of who agreed to transfer was 29 to 1. If we are considering policy issues, as opposed to engaging in a scholarly discussion, that is the most important metric for determining whether the teacher transfer incentives could be scaled up. The TNTP is proud of its supposed ability to look down on the teaching profession and micromanage the nation’s schools. Not even they can believe, however, that reformers can find enough of the 3% of top teachers who are willing to join their crusade and persuade them to staff struggling schools.
The TTI’s good news is that elementary student performance increased by 0.10 STD in math and 0.25 in reading. They estimated the gains would translate into a 4 to 10 percentile improvement, which would be fantastic -- if more than a a few dozen top teachers could be found and persuaded to move and get those gains.
About 1/8th of these elite teachers left their more challenging schools by the end of the first year. They had a higher than average retention rate for the first two years, however.
The bad news is that middle school scores did not increase. There was little sign that the top transfers and the top remaining teachers were able to change school cultures. Moreover, after the top transfers collected their incentives, they were just as likely to leave the low-performing schools. In other words, such a program could not be scaled up unless the entire process was continually repeated.
In a rational world, reformers would prioritize. They would not invest so much in seeking a newer form for monetary incentive claiming, against all evidence and common sense that it can be scaled up. They have committed so much of the political capital to value-added evaluations, however, and they are guaranteed to worsen the attrition of teacher talent in the schools where it is harder to meet test score targets. Apparently, they need to still believe - or claim to believe - that performance pay can prevent the exodus of teachers from the toughest schools. At some point, however, they should realize that similar to other market-driven reforms, like the college readiness experiments and choice, this approach is likely to continue to inflict further damage on the students and educators in the poorest schools.
What do you think? Were it not for test-driven accountability, could financial incentives be part of a comprehensive effort to improve poor schools? Would carrots have made sense if value-added sticks were not such a danger? Regardless of whether bonuses are involved, why can’t reformers see that the way to improve low-performing schools is improving the conditions in low-performing schools so that high-quality teaching and learning occurs?
John Thompson was an award winning historian, with a doctorate from Rutgers, and a legislative lobbyist when crack and gangs hit his neighborhood, and he became an inner city teacher. He blogs for This Week in Education, the Huffington Post and other sites. After 18 years in the classroom, he is writing his book, Getting Schooled: Battles Inside and Outside the Urban Classroom.