These methods involved a lot of change at the schools involved, including changing a number of principals and teachers. But the same student body that had been dramatically underperforming was no longer doing so. Fryer draws the hard lesson explicitly. We know many of the changes that can be made to improve low-performing schools dramatically within a few years. The financial costs of these changes are manageable. But the school systems that need to be changed, and many of the people currently working in those systems, are not ready to make the needed changes.
I remain skeptical. I continue to believe in the ultimate triumph of the null hypothesis. But Fryer is a careful, credible researcher.
I’m a fan of Fryer’s “experiments” approach to schools. But I wonder if perhaps his results are undermined by a key weakness: the schools were not randomly selected. Fryer needed some degree of cooperation from school boards and parents (didn’t he?). Thus, the results should be interpreted as treatments for schools that were “compliant”, and the measured improvements, upward-biased. No?
If the null hypothesis is eventually rejected, then does what is learned in school matter?
By the way, is one interpretation that schools actually do change but they have just already made most of the easy changes?
And is that the same thing as the null hypothesis? I think it is weird to be doing something to everybody that simply can’t be improved. I’d easily buy that it isn’t feasible to be improved.
Mr. Fryer is also an excellent public speaker. The talk is well worth listening to.
I like Fryer, but apart from tutoring, none of those features would have had a dramatic impact on test scores. And tutoring has a great impact on test scores, but I haven’t seen any evidence of its impact on long-term achievement. What tutoring does is give low incentive kids a reason to pay attention long enough to learn something.
BTW, the fact that he used HCZ as a model and cites its great test scores is problematic, too. HCZ rigs the game by kicking out classes that don’t show improvement, and has 30-40% attrition rates. So these things don’t really “work” for HCZ, either.
My theory: teachers stressed test scores, kids got some close attention in high school math, and in math, the kids had some upward ability that up to that point hadn’t been captured on tests.
In reading, on the other hand, all the kids ability was being captured on tests already, so no upward movement.
We tend to use test scores as a proxy for ability, but I suspect that charter school are pushing the limits of that proxy. Best example: the complete lack of a charter school bragging about its ACT or SAT scores. If KIPP kids were getting higher scores after a great start in middle school, if HCZ kids were hitting even an average of 550 on the SAT per section, we’d be hearing of it. We’re not.
It seems like a non-starter for public schools. Reread the first sentence: “These methods involved a lot of change at the schools involved, including changing a number of principals and teachers.”
What teachers’ union is going to approve such a change?
Re the null hypothesis, do you intend it to mainly apply to public schools only, or more broadly to the education of youth and to ongoing education? I believe it for the public schools, due to entrinched interests. However, for the broader question, I believe there are lots of ways to make education work better, and that market solutions can find them. Indeed, I believe the Internet has already vastly improved general education in the last decade, through services like Wikipedia, StackOverflow, and Youtube.