Charter Opponents Running Out of Excuses: Student Gains Continue to Withstand Scrutiny
The flag flies at half staff in front of the Neighborhood House Charter School in the Dorchester neighborhood of Boston, Tuesday, April 16, 2013. (AP Photo/Michael Dwyer)
How is it possible that urban charter schools are increasingly high-performing? “Teaching to the test,” critics will say. But that claim is being debunked by serious research.
A newly published study shows that charter students enjoy greater achievement across the board, posting significant gains on both predictable test questions as well as those that their teachers could not have prepared them for directly.
A growing body of evidence shows urban students make substantial academic gains when they are lucky enough to enroll in a charter school instead of their assigned district school. Their critics like to claim that these gains are an illusion. One prominent argument is that charters only seem to yield gains in student achievement because they focus heavily on test preparation, not real learning.
A new study by Columbia University professor Sarah Cohodes, published in the latest edition of the academic journal Education Finance and Policy, presents compelling evidence to support what anyone who has walked through a high-quality charter school already knows to be true: The gains made by charter school students are real, and not driven by “teaching to the test.”
Cohodes uses data from enrollment lotteries in Boston to measure the effect of charter schools on student performance. The benefit of using enrollment lotteries is that, since a coin flip essentially determined whether a student was offered a seat in a charter, we can be confident that the difference in outcomes between those in the sample attending a charter school and those attending a district school are not driven by unobserved factors related to admission. Any differences can credibly be assigned to charter school attendance.
Use of enrollment lotteries to estimate the effect of attending a charter school isn’t so novel. Prior research using an identical method has found that Boston’s charter schools produce enormous improvement on average test score growth for the overwhelmingly low-income and minority students who attend them.
The new study’s contribution is to evaluate the impact of attending a charter school on student performance by the type of test question. There are categories of questions that are consistently found on a particular exam. Other questions are used for calibrating the test for the following year. And still others test concepts that only show up on the exam every once in a while.
Students attending Boston’s charter schools did much better than their counterparts in district schools on both the obscure and the predictable questions. They also performed similarly on open-ended questions and on fill-in-the-bubble multiple choice questions. These students, in other words, showed the same relative gains on questions that could not be “taught to” as they did for those that could.
These results strongly imply that the measured effect of charter schools on academic achievement represent true learning gains.
Charter school effects are not caused by smoke and mirrors, as some of their staunchest critics continue to attest. One by one, the arguments against expanding urban charter schools are falling away—at least that’s the case for those who value evidence over anecdotes. It is becoming quite clear that urban students truly benefit from attending effective schools offered by the charter sector. Cities like Boston with effective charter sectors can continue to foster the growth of these independent schools of choice without fear that they are being duped by test manipulation.