Education

Those are not-so-useful stats.

I know charter schools that are great for kids' education. I also know charter schools where the kids are warehoused. They learn nothing.

Far more of the latter than the former.

But while they carefully matched students for income, etc., that's probably not all the relevant criteria. A lot of kids are put in charter schools because they are singularly unmotivated to do squat. Those who are doing reasonably well are more likely to be left in public schools. So the data are skewed and the overt sincerity of the attempt to show that the data *aren't* skewed distracts you from seeing the relevant criterion.

There's a craft, a science, and an art to establishing default hypotheses and figuring out what characteristics to control for. Many ed faculty come to a problem already knowing what's important. That keeps them from examining the data and seeing what's actually going on. (Yet they're all over the idea that teacher candidates must chuck any preconceptions that they know what it means to be a good teacher. Look only at the data, as interpreted by the pundits and their little altars.)

Yet that doesn't even do justice to what's going on, as far as I've seen it. I've seen kids put in charter school mid-year, in time to salvage their fall grades. They are given passing grades in a charter school instead of failing pubic school. Then they transfer back in the spring and bring with them a good average that we have to accept. They fail every test, their average goes down from the minute they appear in the public school classroom again, and they squeak by with a barely passing grade.

The public school's graduation rate's saved from another ding, another kid graduates with a degree that really means little once it's gained. Of course, those students who can't pull this trick off are left to fail in the charter school until the following fall, when they often repeat the year or drop out.