Michael Birnbaum has an article in the Washington Post today about Maryland educators’ resistance to a proposed evaluation scheme put out by the Maryland Board of Education. The new rule would require that at least 50% of a teacher’s evaluation be based on student performance, as measured by standardized tests “and other measures.”
Apparently, the goal of the Board is, at least in part, to chase after Obama’s “Race to the Top” dollars.
Now, I adore President Obama as much as the next raving liberal, but I’m crushed that someone so apparently intelligent has bought in to the “merit pay” nonsense that is sweeping the nation.
The idea of basing teacher pay on student performance is ridiculous. Teacher pay should be based on teacher performance: are teachers observing best practices? Are they teaching well?
Student performance is determined by so many factors outside a teacher’s control that basing teacher pay on said performance is simply unfair.
Writing in Economics of Education Review in 2010, Lucia Tramonte and J. Douglas Willms of the University of New Brunswick observe:
Most studies of social mobility have found that academic achievement and occupational attainment are largely determined by people’s family origin and educational experiences. These studies have focused primarily on the roles of socioeconomic status (SES), family structure, and family resources, including economic, cultural, and social capital.
The idea that teachers are not solely – or even largely – responsible for the average academic performance of their students might seem controversial, but that’s only because our society has been brainwashed by the mythology of Jaime Escalante — the notion that a brilliant teacher, through heroic feats of self-deprivation and superhuman dedication, can transform the most troubled students into Harvard neuroscience graduate fellows with only two months’ time and a fresh pencil sharpener.
This is of course nonsense.
Let me be clear: my argument isn’t that good teachers (and bad) don’t make a difference. My argument is that, for student bodies as statistical models, individual teachers have extremely limited influence when compared with other factors, particularly family and socioeconomics. That means that holding teachers “accountable” for student performance is ridiculous. Teachers should be observed and evaluated by experts in their fields and held accountable for the quality of their classroom practices.
But basing teacher evaluations on standardized tests is both cheaper and easier, because then you don’t have to actually find experts or make them do observations.
But the biggest problem with basing teacher evaluations on student performance isn’t the unfairness – it’s the impact it will likely have on education for the most marginalized students.
It is an inescapable statistical reality that variation in standardized test scores is attributable almost entirely to differences in socioeconomic status, with poorer students having more trouble.
If teachers are going to face reduced pay or job loss as a result of poor student performance, how on earth are we to convince the best teachers to take jobs in the toughest schools?