Here's a reaction to my special "Stupid in America," which aired this weekend:
Dear Mr. Stossel,
Do you understand how totally unsupported [your] claims are? All the studies of charter [schools] show they do worse on average than public schools, with a few cherry-picked examples doing better but most doing worse.
All the studies? That email came from a public school teacher in Virginia. And I can see where he gets such ideas. In 2006, the U.S. Department of Education announced the results of a big study: "Children in public schools generally performed as well or better in reading and mathematics than comparable children in private schools." The media ate it up. The New York Times put the study on its front page, along with a quote from teachers' union president Reg Weaver, who claimed it showed "public schools were doing an outstanding job."
It turned out that the Education Department's researchers had to torture the data to conclude that public schools performed well. The actual data showed that charter school kids scored higher, but then the researchers put test scores "into context" by adjusting for "race, ethnicity, income, and parents' educational backgrounds to make the comparisons more meaningful."
Maybe it's unfair to call that "torturing the data." Regression analysis is a valid statistical tool. But it's also prone to researcher bias, and there's plenty of that in the government monopoly.
The Department of Education's researchers acknowledged their method could be flawed: "Ideally, to ascertain the difference between the two types of schools, an experiment would be conducted in which students are assigned [by an appropriate random mechanism] to either public or private schools." The NY Times didn't find it "fit to print" that.
Last year, the Education Department came out with the results of one of those experiments - a "gold standard" study that compared students who were accepted via lottery at 28 different charter schools to students who applied to the same schools but who were randomly rejected in the lottery.
They found no significant difference in test scores, but parents were almost twice as likely to be happy with the charter school -- 70% of lottery winners said their school was "excellent," compared to 38% of lottery losers. Kids liked the charter schools more, too, 75% to 62%.
The study found that some schools drastically increased kids' scores, but that others did worse than the public schools. Market competition means that, over time, good schools will expand while the bad ones die.
School choice opponents also claim that successful charters "cherry-pick." Their students get higher test scores because charters enroll fewer special-needs kids, and kids who speak little English. Such "cheating" would be difficult since admission is done by lottery, but it's possible. Maybe the indifferent parents and parents who don't speak English don't enter the lotteries.
I challenged successful charter directors Eva Moskowitz and Deborah Kenny about "cherry-picking". Kenny called it a "myth" and Moskowitz told me the lottery is totally random, and that 70% of eligible kids apply for her schools (on top of that, 78% of the kids that attend her school are eligible for free/reduced school lunch).
Former DC schools chancellor Michelle Rhee says charter school success is not about charters picking their students. "I'd have a school where 10 percent of the kids were on grade level. And right down the street I would have a KIPP school (where the kids are picked by lottery) and 90 percent of the kids were proficient... They are getting kids in who are scoring very low in proficiency, and by the time they leave they are achieving like suburban kids are."