In an October 2012 preliminary working paper, Ron Zimmer of Vanderbilt and Cassandra Guarino of Indiana University consider whether public charter schools “push out” low-performing students. The idea is that charter schools might try to improve their performance by “educating fewer challenging students.”

But, as the authors say, “the claim of ‘pushing out’ low-achieving students has consisted largely of conjectures on websites or opinion pieces.” Using data from “an anonymous major urban school district with a large number of charter schools for the school years 2000-01 through 2006-07,” they “examine whether low-performing students in charter schools appear to exit at higher rates than those in traditional public schools (TPSs).”

They conclude:

our descriptive results suggest that students exiting charter schools do have slightly lower achievement levels than their former peers. However, the same holds true for TPSs. When examining these patterns with formal regression models, we find no evidence that low-performing students are more likely to exit charter schools than above-average students. In fact, we find some evidence that low-performing students are more likely to exit TPSs than charter schools.

It would appear that, as the authors put it, “the ‘push-out’ hypothesis is a weak argument to use to oppose the establishment of charter schools.”

We wrote about other studies of charter school effectiveness in this policy brief on Initiative 1240, which would allow some public charter schools in Washington.

(The link to the paper was via Modeled Behavior‘s twitter feed.)