Malcolm Kirkpatrick
Banned
- Joined
- Jan 20, 2006
- Messages
- 4,046
Doesn't work.The OP.What "more recent data"? Link=___?After some digging I did find them. The first youNeither condition describes the work I reference above. The random-assignment voucher lottery is as close to a double-blind trial as you'll get in education research.I'd like to see how they controlled for confounding factors. I see this argument sometimes with how prevalent 'private education' is in Japan, ignoring the fact that much of that is in the form of extra education and high-end schools from high competition and social emphasis on one.I have read otherwise.Several studies have shown that students at private schools perform no better than those at public schools once you adjust for socioeconomic differences of the students, so it's not a big surprise that charter school students don't perform better than those at traditional public schools.
Gerard Lassibile and Lucia Navarro Gomez,
"Organization and Efficiency of Educational Systems: some empirical findings"
Comparative Education, Vol. 36 #1Joshua AngristMaterial deleted...
"Randomized Trials and Quasi-Experiments in Education Research"
NBER Reporter, summer, 2003Herman Brutsaert compared student performance in government and parochial schools in Belgium (which subsidizes parents' choice of school) and found (a) higher mean scores in parochial schools and (b) a lower correlation between parent income and performance in parochial schools (i.e., government schools exacerbate inequality).Material deleted...
Multi-country comparisons of independent and government schools consistently find a private school advantage both in performance and cost.You can see for yourself that these studies address the question. In any case. I gave the references to two (Lassibile and Gomez, Angrist. et. al) so you can check. I mention also Herman Brutsaert, but you'll have to search with that name, since I don't recall the publication. See also Lockheed and Jiminez (referenced above) and various studies of the Chilean voucher policy. Andrew Coulson pointed out a defect of studies that "control" for parent income: they use "free and reduced lunch" as a proxy for "poor". Government schools use this classification. If a parochial school does not offer government-subsidized lunch and charges one price, it will have no use for the classification. All such a school's students will be classified as "non-free/reduced price lunch", whatever their parents' income. Caroline Hoxby has pointed out a second difficulty with studies that compare charter schools to the NEA/AFT/AFSCME cartel's schools (the "public" schools): the proper comparison is to the schools that charter kids would otherwise attend, not to all government-operated schools.In short I'm questioning the applicability of your citations here as they have been so misapplied in the past.linked toreferred to is far to narrow a focus and isn't even attempting to address what you claimed, and thus I would call it not applicable when coupled with the age of the research. Your second is highly criticized, but that doesn't make it invalid. However, I'm certainly not going to at this point trust a study whose very methodology is under fierce debate at least until that's ironed out. At any rate, neither invalidates the more recent data.
1. The OP has links to news articles about charters. To say that the NBER and World Bank comparisons of private and government schools don't "invalidate more recent data" is like saying that decades-old measurements of Dead Sea salinity don't invalidate more recent studies of fertility in wild populations of howler monkeys. Duh! However, to the extent that statistics on private schools relate to charter schools, they provide support for two generalizations:
a) As institutions displace individual parents in the power to determine for those parents' children the choice of curriculum and the pace and method of instruction, overall system performance falls.
b) Political control of school harms most the children of the least politically adept parents.
2. You call news articles "data" and question peer-reviewed random-assignment studies? Interesting preference.
Someone call the ******* Mayor! This is such a shocking revelation!
