So, here's reformy argument #1: we must, must, MUST test the crap out of the kids, because...
Research over the last two decades has confirmed what most parents already knew: Teacher quality is any public school’s most important asset. Taking that simple and obvious premise seriously means working to identify and remove ineffective teachers. A bipartisan group of lawmakers in New Jersey and nationwide is pursuing this path. [emphasis mine]This is a variation on a common reformy theme: "Teachers are the most important in-school factor in student achievement." It seems that everyone says it: Education journalists. Professional union busters. Corporate edu-vultures. Presidents. Maniacs who happen to be the Secretary of Education. ACTING state education commissioners. Governors who yell at teachers on YouTube. Mouthpieces of autocratic mayors.
And even, lord help us, some teachers themselves. All agree: "Teachers are the most important in-school factor in student achievement."
Let me ask all of these fine folks one question: if you were looking for students, where would you go? A school, right?
Well, by a huge factor, the most important determining factor in student achievement is the difference in student characteristics. The student's family environment and the student's qualities determine about 60% of the differences in student outcomes. About 20% is attributable to the school; around half of that is the teacher. And 20% is just error.
The sad fact is the correlation between poverty and student outcomes is nearly perfect.
Does this mean that poor kids are doomed to failure? That good teaching doesn't matter? No, of course not. There are many things we can do in our schools to help mitigate against poverty, starting with adequate school funding. Teacher quality does vary and we ought to ensure policies to get the best teachers in front of our classrooms.
But, granting all that, Winters's argument here fails on two levels. First, there is no evidence that basing teacher evaluations on bubble tests will improve teacher quality; sadly, the anecdotal evidence is pouring in that it will probably do just the opposite.
Second, since Winters overstates the teacher's effect on student learning, he ignores the one critical thing we must do to raise student achievement: close the poverty gap. Winters gives away the game with this statement:
Anyone who understands the importance of education won’t be surprised to learn a recent study by economists at Harvard and Columbia universities showed that assignment to one or another teacher is related to later life outcomes, such as the likelihood of early pregnancy, the chances of college attendance and lifetime earnings.Yet another misreading of Chetty et. al. (more on this later in the series). But aside from that, notice how Winters gets it backwards: he thinks we can close the poverty gap by closing the achievement gap. He mistakes causation for correlation: he thinks low achievement is causing poverty.
This defies all common sense. Is Winters really prepared to say that the reason Chatham and Scarsdale and New Canaan and Grosse Point and Los Altos Hills have high-achieving students is because they have better teachers? And the primary difference in the lives of children in Newark and the South Bronx and Hartford and Detroit and South Central is that their teachers suck?
Let's look at Bruce Baker's table one more time:
The evidence is crystal clear: the most important in-school factor affecting student achievement is the difference in student characteristics. Stop the weasel words already.
Students First's new spokesman: "Teachers are the most important in-school factor!"
ADDING: Extra credit reading.
ADDING MORE: Even more via Paul Thomas.