`I couldn’t afford to learn it.’ said the Mock Turtle with a sigh. `I only took the regular course.’
`What was that?’ inquired Alice.
`Reeling and Writhing, of course, to begin with,’ the Mock Turtle replied; `and then the different branches of Arithmetic– Ambition, Distraction, Uglification, and Derision.’ 

                                                                                                                                    – Lewis Carroll

A couple of days ago the South Charlotte Weekly carried an article that focused on protest by parents against a program that uses the results of standardized tests taken by students to determine teacher compensation.  The interesting thing is, as one of protestors pointed out, is that similar programs have been tried and abandoned in Texas, Tennessee, California and New York. Teachers in the community said they were not against the idea of merit-based pay, but that they felt that the results of standardized tests do not constitute a judicious measure of a teachers’ ability or dedication to the job.

I think this is a valid argument. If you put in place a system that rewards teachers who train students to perform well in standardized tests, then the public school system if run efficiently will produce students who perform well in standardized tests. If this was indeed the only skill required for success in later life, all would be well. I think you would be hard pressed to justify such an assumption, though. It is obvious even to those unacquainted with the science of pedagogy that teachers should be adept at nurturing a wider and more  complex set of skills than merely the ability to perform well on standardized tests.  Therefore the attempt to ‘rationalize’ public education through the use standardized test results to determine teacher remuneration logically leads to a sub-optimal result.


So, the ‘test’ is inefficient, but does that mean the idea of performance-based incentives for teachers is flawed? Surely one could combine several measurable criteria in order to come up with a more accurate way of determining which teachers should be rewarded, thus creating the right incentives for a more successful public education program? Well unfortunately it appears that it isn’t quite that simple. Linking teacher compensation to student performance, no matter what the chosen performance indicator is, does not seem improve student performance.  Roland Fryer of Harvard spear-headed a study for the National Bureau of Economic Research examining the effect of teacher incentive programs on student performance across two hundred schools in the New York public school system, where each school was allowed to determine its own student performance indicator. The study I found NO evidence to support the fact that teacher incentives improved student performance. In fact there was evidence to suggest that teacher incentives may negatively impact student achievement, especially in larger schools. Depending on how the criteria are structured, student performance based incentives can either cause teachers to concentrate on median student performance, reducing time and effort spent on students either too far below or above the median, or cause teacher to concentrate on the high performers, neglecting the rest of the class.

Tarun Jain and Tulika Narayan used the results of behavioral experiments conducted in India to examine the challenges of designing an optimal incentive schemes for schoolteachers. They found that in environments where there are social prejudices and hierarchies, incentive based schemed coordinate teacher effort away from low-status students and toward high-status students. Thus merit-based pay can create incentives that promote social discrimination.

It seems that introducing market-like incentives into the public education system seems to distort incentives. I wonder if anyone else finds this ironic.