Count on Jersey Jazzman to catch the flaw in the CREDO study of charters in his home state.
It purports to study schools but in fact it compares matched students, not taking into account the peer effects.
Like, if you go to a school that has weeded out or excluded the troublemakers, you learn more. Where do we go with that?
Diane Ravitch's blog 
Jersey Jazzman does point to a big issue in education. Jersey Jazzman argues that If Diane were to go to a charter school with Michelle, Michelle would get a better (at least measured by the metric used in the CREDO study) education. It would seem that if we want Diane to get the best possible education, she should be encouraged to go to whatever school is filled with students like Michelle. This does contradict other goals we have for education in the US, and we need to think carefully about this.
LikeLike
TE, you hit the nail on the head:
http://jerseyjazzman.blogspot.com/2012/07/new-segregation-in-new-newark.html
The fact is that we already segregate our kids in our public schools by class and by race. Liberals like myself have to come to terms with this. If we think the segregation of students in cities by class is unacceptable in a charter model, how can we acquiesce to it in the suburban-urban segregation the happens today (and is particularly awful in my own state)?
My main objection right now, however, is that we are not having an honest debate about what is happening. Corporate reformers fool themselves into thinking KIPP or Noble or Success has a “secret sauce.” They don’t: they’re just doing what the ‘burbs do on a smaller scale.
This is an incredibly dangerous position to take, and I believe it is fueled by an exterior motive.
LikeLike
Good stuff. Several thoughts:
1. Peer effects are the achilles heel of all of these school-level studies, even the “gold standard” random assignment studies that track kids that get into schools via lotteries versus those that “lose” the same lottery.
2. It would be interesting to see studies that try to estimate the potential size of the peer effects and adjust results accordingly. I’m not sure if that is possible to do, though.
3. Because of peer effects, it might be better to look at community or district-level studies. As an example, Eva Moskowitz (!) wrote an editorial last month that pointed out that Harlem as a whole has gone from the bottom of the pack to the middle of the pack on the state exams: http://goo.gl/EEaGn. New Orleans, of course, in which almost all of the schools are charter schools is another interesting example to study.
4. I agree strongly that there is a hypocrisy here to the extent that some are suggesting that poor parents can’t even select lottery-based schools in their communities while non-poor parents can select schools via geography, testing, money, etc. However, this is a different issue than how we should study the performance of schools. Peer effects are likely important.
5. I’m not sure why people aware of peer effects would then conclude that high-performing charter schools are, therefore, just the same schools with different kids. That sounds like a unscientific conclusion. Many people that I know that love these charter schools are most impressed by what they see and learn when they visit the schools. The problem of peer effects only motivates that school-level studies are very difficult to draw conclusions from.
LikeLike
There have been a number of attempts to estimate the size of peer effects on learning. The Handbook of the Economics of Education has chapter entitled “Peer Effects in Education: How Might They Work, How Big Are They and How Much Do We Know Thus Far?” The abstract suggests that there are large peer affects for high ability students, racial composition and achievement of same race peers are often found to be important.
Here is an older NBER paper about Peer effects: http://www.nber.org/papers/w7867
LikeLike
Thanks! I wonder if these estimates could be reasonably reflected in school comparison studies to try to ascertain if charter schools outperformance could be explained completely by peer effects. That sounds challenging! In any case, I’ll take a look at the article.
LikeLike
Jersey Jazzman belongs on your Honor Roll!
LikeLike
I consider it an honor every time Dr. Ravitch posts on one of my pieces. It’s an honor to be mentioned on this blog at all considering the great writers regularly featured here.
As always, Galton, thanks for your support.
LikeLike
The press release about NJ Charter school success had intentionally omitted the fact that ELL students and Special education students represent less that 1% of the total number of students enrolled in charter schools. These kinds of prepared press releases serve the purposes of the Governor and the Commissioner of Education. The lack of transparency serves to diminish their integrity and credibility in matters of education.
LikeLike