New York mayor Michael Bloomberg, New York governor Andrew Cuomo, and New Jersey governor Chris Christie are the latest high-profile policymakers to pursue radical changes in the evaluation of public school teachers. While the policy details—in particular, how they would use students’ test scores to evaluate teachers—get the most attention, the most promising aspect of these proposals is that they represent a fundamental shift in philosophy. The new approach is much better aligned with what we now know about teachers and how we should employ them.
Under the current system, public schools measure teacher quality by placing heavy weight on a set of professional credentials. In order to enter the classroom, a prospective teacher must earn a degree from an education college and must be certified. Teachers earning advanced degrees get rewarded with higher salaries.
The teacher’s actual performance in the classroom, however, escapes serious scrutiny. Officials base teacher evaluations primarily on classroom observations by principals that take place usually only once or twice a year, and often for less than a full class period. Typically, 98 percent or more of teachers in a school system are identified as “satisfactory” or better.
The philosophy underlying this system appears sensible at first glance. Public schools rely on the screening process to ensure that only teachers meeting minimal standards are hired. They then provide incentives for teachers to obtain professional development (advanced degrees) that will, or so the thinking goes, improve their classroom performance. Since all public school teachers receive similar training, teacher quality shouldn’t vary much, making the evaluation system of little importance.
The problem, however, is that over the last two decades, empirical research has revealed that teacher quality does in fact vary substantially, not only across but within public schools. And for an individual student, the difference between being assigned to one or another teacher can mean as much as a grade level’s worth of learning during the school year. Further, recent research by economists at Harvard and Columbia shows that teachers influence their students later in life: being assigned, that is, to an effective versus a less effective teacher has a bearing on the likelihood of teen pregnancy, on the probability that a student goes to college, and on students’ earnings as adults. The enormous disparity between the best and worst teachers indicates that the current system does a poor job of ensuring teacher quality.
Research also shows that the credentials prized under the current system tell us next to nothing about how well a teacher performs in the classroom—and they explain only about 3 percent of the variation in teacher quality. Obtaining a master’s degree, it turns out, is simply unrelated to a teacher’s effectiveness. The positive results of alternative-certification programs, such as Teach for America, illustrate that great teachers don’t need to graduate from an education college.
While differing in some specifics, all of the recent proposals to revamp teacher evaluations would turn the current system on its head. Rather than screening teachers before they’re hired and then assuming that they’re performing well, the new strategy focuses on distinguishing between the best and worst teachers by assessing their actual classroom performance.
The current system might have made intuitive sense in the past, but modern research findings make its continuation indefensible. We can’t go on pretending that all teachers are effective when both common sense and empirical research tell us otherwise. The recent movement to revamp teacher evaluations could bring us closer to the day when all public school students have the chance to be taught by high-quality teachers.