How Charters Fake Success

Jersey Jazzman is one of the premiere edubloggers out there, and I rarely mention him here because I generally don't have anything to add except, "Yeah. What he said. Read that." But his post from yesterday about a subtle way in which charters can cook the books is extra worth taking another look-- and I think I can add some value for the stats-impaired readers.

I'll start by echoing his point that all charters are NOT money-grubbing scams, and that a well-done charter school can be a great addition to a school system.

But some insist on trying other routes to success, some more subtle than others.

The technique that the Jazzman lays out is more subtle than, say, paying yourself rent for the building. But here's how you do it:

1) Open up charter in urban high poverty area.
2) Accept only a small percentage of high poverty students
3) Compare your low/high needs student ration to the state, rather than the local district
4) Compare your results to the local district rather than the state

The Jazzman lays this out meticulously with charts and data. But some folks don't speak statistics fluently (I know, because I live with one of them), so let me see if I can turn this into an analogy.

Let's say that in our state of Curmudgistan, low ability, high poverty students are blue, and high ability, low poverty students are red. If we look at the state as a whole, the ratio is about 1:1 (the figures in this example are all manufactured for effect, not perfect accuracy). So my state, on average, is a lovely purple.

But in the urban center of Grumpville, the ratio is more like 3:1. Grumpville is just a slightly violet shade of blue.

So I start Grumpville Academy. I recruit high income students like crazy, and I take a small number of high needs, high poverty students. My ratio is 1:1. "Look!" I declare. "We're not skimming at all. We have the same numbers as Curmudgistan."

And then when my results come in, I declare, "Look! We are a lovely purple, while the rest of Grumpville is pretty much blue. Clearly we are a far better school that the public school system of Grumpville. Yay, us!"

Two things to note about this:

1) The results are not scaleable to all of Grumpville. Because the ratio in Grumpville is 3:1, and my success is based on a 1:1 ratio. Put another way, if I want to make purple, and I've got 300 gallons of blue and 100 gallons of red, I either have to go to the store and get more red, or I have to throw away 200 gallons of blue.

2) The mix matters. This is a subject that deserves its own attention, but school culture and climate matter. If I take a cup of blue, what happens next to it depends on whether I pour it into a vat or more blue, or into a vat of red. We don't talk nearly enough about students in this context, and instead keep insisting that the school culture can be completely controlled by teachers an administrators who are, in my analogy, the buckets.

Again-- to get a more data-y and smart explanation of all this, go read the original Jazzamn post (and then read the rest of his blog as well).