Recently, CREDO published a rigorous study on Ohio charter school performance. Here is what they found for Ohio’s biggest cities:
In sum, charter schools in Ohio’s major cities are doing roughly the same as traditional schools.
Most reform advocates, including myself, view these results as incredibly weak. Generally speaking, I’m not enthusiastic about a charter sector until I see results over .05 effect size that are ideally trending toward .1 to .2 effect sizes.
That being said, if you take the position that traditional school districts are doing a pretty good job, then, to be consistent, I think you also need to give the Ohio charter sector a pat on the back. After all, they’re basically keeping up with the traditional system.
So how come the defenders of traditional systems aren’t applauding the performance of Ohio’s urban charter schools?
NK-
Paul Petersen noted some considerations about these Credo charter studies that have never received the attention they deserve: basically that there are some significant factors that have not been taken into account such as the age of the schools and the length of attendance. We know that kids take a temporary academic hit when they transfer into a school as the students adjust to a new school and social setting. We also know that new schools are never at their best during their shakedown period. Thus charter sectors heavily populated by brand new schools with kids who just transferred in are going to be fighting with one arm tied behind their back in the short run. Writing for the WSJ in 2010, Peterson noted that Credo’s work would be more informative in the future when they could present data for charters that have been up and running for a number of years, and pointed readers to the more rigorous random assignment studies:
http://www.wsj.com/articles/SB10001424052748703909804575123470465841424
Thus for example Credo’s 2009 report panned Florida charters:
http://credo.stanford.edu/reports/statepressreleases/Florida.pdf
But a sophisticated regression discontinuity analysis found Florida charter students were more likely to graduate from high school, attend college, persist in college and had higher earnings at age 25:
http://cesr.usc.edu/documents/charter_high_school_effects_on_attainment_and_earnings.pdf
This is not to say that everything is great in Ohio charter land (I would wager my left foot that it is not) just that the Credo studies should be understood to have limitations that keep them a notch below random assignment studies in rigor.
LikeLike
Matt, good to hear from you.
Two thoughts:
1. I agree with length of school opening. Only counting charters that have been open 3+ years could be useful.
2. I’m a little worried about student length in schools, because selective attrition could play a role here, especially if this metric became a more prominent marker for equality.
-N
LikeLike