Incentive pay for teachers—why it’s a really dumb idea that needs to be abandoned now The idea that paying teachers bonuses if their students do well on tests has been gathering momentum recently, which is alarming because it is an idea that will cost a lot of money to implement, cannot be done fairly, doesn’t work, and has the potential to lower student achievement. First, despite the claims of the value-added proponents, we cannot divide up increases in student learning and allocate them to individual teachers. As Jesse Rothstein’s work shows, good teachers benefit their students for at least two years after they have stopped teaching them, and conversely, the total harm caused by bad teachers takes years to materialize. Second, we cannot use classroom observations to work out who the good teachers are either. While the work of the Gates’ Measures of Effective Teaching program, and the work of Consortium on Chicago School Research, has begun to tease out what teacher behaviors are associated with increased student progress, they are still accounting for less than 20% of teacher quality. So if we pay bonuses to teachers who rate highly on one of their observation protocols, we do know that, on average, the bonuses will go to teachers who are more productive, but the differences are small. And then there is a real danger that teachers who are currently highly effective in ways that are not represented in the framework will become less effective because of the incentive to ape the protocol in order to get the bonus. Third, a recent study showed that even $15k incentives weren’t enough to raise students’ scores, which suggests that teacher effort is not the problem. And from what we know about motivation (see Daniel Pink’s book “Drive”), performance-contingent rewards tend to lower performance on high-complexity high-creativity tasks. So what should we do? Pay teachers fairly, and let them get on with it. Create an expectation that every teacher should improve their classroom practice every year, not because they’re not good enough but because they could be even better. And create support systems that support each teacher in lifelong improvement.
Dylan Wiliam’s post on Incentive Pay–interesting reading December 4, 2011