Relevant data

I find Matt Yglesias‘ interpretation of Erik’s initial post on education reform exceedingly uncharitable, to say the least. But the crux of the issue seems to be this:  Matt is suggesting that Erik is underestimating the importance of teacher quality in educational outcome, and Erik is suggesting the opposite of Matt’s evaluation. Fortunately, we don’t need to rely on estimations. In her excellent review of the appalling Waiting for Superman, education policy expert Diane Ravitch gave an account of the relevant research:

The movie asserts a central thesis in today’s school reform discussion: the idea that teachers are the most important factor determining student achievement. But this proposition is false. Hanushek has released studies showing that teacher quality accounts for about 7.5–10 percent of student test score gains. Several other high-quality analyses echo this finding, and while estimates vary a bit, there is a relative consensus: teachers statistically account for around 10–20 percent of achievement outcomes. Teachers are the most important factor within schools.

But the same body of research shows that nonschool factors matter even more than teachers. According to University of Washington economist Dan Goldhaber, about 60 percent of achievement is explained by nonschool factors, such as family income. So while teachers are the most important factor within schools, their effects pale in comparison with those of students’ backgrounds, families, and other factors beyond the control of schools and teachers. Teachers can have a profound effect on students, but it would be foolish to believe that teachers alone can undo the damage caused by poverty and its associated burdens.

So here we are. Existing research suggests that somewhere between 7.5–20% of outcomes variable by teacher quality, 60% to non-school factors. These studies, Ravitch notes, include those published and promoted by those in the reform movement.  So once we’ve got that sorted out, hopefully we can all agree that a policy measure that improved teacher quality could potentially make a positive impact, but much less of an impact that improving ‘nonschool factors’ related to poverty.

But I suppose if attending to those nonschool factors are off the radar, what can we do to enhance teacher quality? The ‘reform’ movement’s answer seems to be, basically, to remove job security and fire teachers based on highly unreliable test scores. As Ravitch notes in this interview, many of these tests were not designed to evaluate individual teachers, and their creators oppose that use.  In order to fire more teachers via a largely arbitrary process, teachers unions may have to be weakened considerably.

Alternatively, Erik and I have an alternative plan: pay teachers well and treat them with respect, and like professionals. This will attract and retain strong teachers and improve morale. The ‘reform’ movement’s proposed solution will fires a few bad teachers and some that are fine, decrease morale (as actual teachers are more acutely aware of the weaknesses of the test scores as an evaluation technique).

Of course, our plan will allow a few bad apples to take advantage of increased pay and autonomy. Insofar as the quality of their instruction is low, and existing mechanisms to improve or remove aren’t able to address this, this will have modest but real negative consequences for students in their classrooms.

But any policy structure is open to exploitation by bad actors in some way or another. My simple, boring proposal is open to mild exploitation by bad or lazy teachers. The ‘reform’ movement’s preferred policy structure is open to exploitation by much more powerful actors, whose capacity to exploit is considerably greater–the testing industry, for one, the Michelle Rhee’s of the world, for another, and of course the grifters taking advantage of the current enthusiasm for charter schools. The latter, in particular, are responsible for some spectacular failures (as Ravitch notes, 37% of charter schools produce worse results that comparable public schools, as opposed to 17% who produce better results), many of which involve paying the CEOs many, many times the annual teacher’s salary while overseeing a school that produces lower outcomes for thousands of students, rather than a few dozen. A teacher cheating the system is frustrating, but the impact is limited. A deeply flawed, failing, profitable charter school cheating the system is a great deal more problematic. Insofar as a goal of a good educational policy structure is to prevent bad actors from taking advantage of the system, the policy reformers tough, hard-headed attitude toward the bad teachers of the world doesn’t count for much compared to the potential bad actors they’d willingly empower and enrich.