I don't want to give away too many of my thoughts here, as I have a major piece coming out this weekend about Vergara in a pretty darn large media outlet here in Jersey (I'm such a tease...). For right now, I'd encourage you to go over to Diane Ravitch's blog: as usual, she has collected some of the best writing on this issue. I think Kevin Welner and Dana Goldstein in particular make some very good points.
For now, I'll just add this:
Judge Rolf Treu's decision reads like a reformy greatest hits list. He's got Hanushek and Chetty, dropping their academic bombs from 40,000 feet, blissfully (willingly?) unaware of the damage they are causing to the teaching profession and the lives they are bound to ruin in the name of "putting kids first."
Treu's use of the Mountain-Out-Of-A-Molehill-Inator, speciously pumping up the rather modest findings in Chetty's study, could have come right off of StudentsFirst's website (and probably did). He bizarrely imagines a public school system where "gifted" junior teachers regularly lose their jobs to "grossly ineffective" senior teachers -- as if this scenario plays out regularly in any place other than Michelle Rhee's fevered imagination.
But what really makes Treu's essay bananas is that it refuses to acknowledge the most obvious thing in the world:
All data from the CA-DOE. Every point on this graph is a California elementary school. The vertical axis shows the average scale score in English Language Arts for Grade 3. The horizontal axis shows how many children at that school qualify for free or reduced-price meals, a proxy measure for economic disadvantage.
If poverty had no effect on test scores, there would be a big blob in the middle of this graph. But look at how closely most of the points cluster around the red line. That line shows a prediction for a school on how well its students will do on the ELA test, given how many kids are economically hurting. That line is telling us that poverty has an enormous influence on test outcomes. And the number in the lower left corner is mathematically confirming this truth.
Nearly 60 percent of the variation in English Language Arts test scores for Third Grade students in California can be explained by the level of economic disadvantage in their school.
If Judge Treu really thinks the California constitution demands that all children must have an equal opportunity to succeed in school, he really has no choice, does he?
If California declares that tenure is unconstitutional, it must also declare that childhood poverty and economic disadvantage are unconstitutional.
Seems to me that Judge Treu, if he's going to be consistent, would have to admit that his decision would be right in line with policies that would tax the crap out of people like David Welch (the money bags behind the California anti-tenure movement), so the state could use the money to better the lives of California's poor, deserving children.
I would very much like to see someone bring exactly this argument before a California court. And I would dearly love it if that court was Rolf Treu's. True tests of intellectual integrity are so rare these days...
ADDING: Some more grade levels:
Are you kidding me? Two-thirds of the variation on California's Grade 5 ELA tests can be explained by free lunch eligibility?! That's extraordinary.
On the CST, the r-squareds run between about .25 and .6, depending on the grade and test. In my mind, we'd expect to see that correlation decrease as students got older, but let's save that discussion for another time...
This summer, I'll hopefully be looking some more at California: it is a truly wacky place when it comes to schools.
I think the only reason the r is lower in the last graph is that the relationship is less linear. It curves sharply upward on the left edge and disperses greatly toward the right. Notice also that the elementary graphs all have huge, dense clusters in the 80–100 %F/RPME range, but in grade 11, there's a noticeable hole between 90 and 100 percent. I'd surmise that represents the poorest students' dropping out of school altogether, which makes it harder to measure test score correlation accurately.
ReplyDeleteThank God for appeals. I didn't follow the trial very closely; do you know who/if the defense brought in as expert witnesses?
ReplyDeleteThis decision is so bizarre, off kilter, Kafkaesque, Orwellian and abusive of normal human logic, it has the same effect as a sucker punch to the stomach. How the hell are these supposedly ineffective teachers getting in the classroom in the first place? They don't hire themselves, they don't observe and evaluate themselves. At what point do the administrators and principals bear some responsibility for these "ineffective" teachers. Of course this is where the billionaire reformers and their obedient flunkies (Judge Treu) blame tenure, seniority, LIFO and unions. Garbage.....top performing states like MASS, CT and NJ have strong unions, tenure, seniority and LIFO. These 3 states are the top scorers on the NAEP tests.
ReplyDeleteFor 27 of 30 years, I taught in California public schools with rates of poverty---as measured by free/reduced lunch/breakfast---of 70% or higher. I retired in 2005 and that number is up to 80% now at the high school where I taught.
ReplyDeleteIn 1994-95, I kept a daily journal of what was happening in my classroom at the high school where I taught, and in the district. Two years ago, I took that journal out of the fire-proof safe in the garage and read through it to see if there was enough material for a book.
That teacher's memoir will be out in a few days. I called it "Crazy is Normal".
I'm sure that it will provide evidence that poverty is unconstitutional.
Good eye, Keith. Dropouts are clearly a factor in this stuff - like I said, hopefully more on this later.
ReplyDeleteMarie, the dribs and drabs are coming out, but apparently Linda Darling-Hammond was there, and Jessie Rothstein, and David Berliner. Up against Raj Chetty and Tom Kane, who Treu apparently fell in love with.
G, the LIFO stuff is garbage, straight up. But I have to give Treu credit: he makes some valid points about the CA laws, which are somewhat incoherent. But -- as you will read Sunday in the Bergen Record -- that could be fixed by making CA's las more like NJ's.
TSV, please, PLEASE let me know when the book comes out. I will make time this summer.