David Leonhardt of the New York Times has written favorably about charter schools, without paying much attention to issues such as attrition and selective admissions. Nor has he explored the impact of charter schools on the public schools that enroll the majority of students or considered the value of public funding of two school systems, one free to choose its students, the other required to accept all. He recently invited charter skeptics to contact him. John Thompson, teacher and historian in Oklahoma, has a message for him.

Feel free to suggest other studies that Leonhardt should read.

Thompson writes:

“In “A Summer Project to Nourish Your Political Soul,” the New York Times commentator David Leonhardt pledges to wrestle with the complexities of immigration and abortion, as well as the issue which he debates most with his readers – charter schools. He’s devoting part of the summer to learning about “vexing issues.” Leonhardt asks “reform skeptics” to “dig into a few of the studies, essays and evidence that have persuaded me.” In return, he promises to keep an open mind when considering our responses.

“The first study that impressed Leonhardt was the Education Research Alliance’s “What Effect Did the Post-Katrina School Reforms Have on Student Outcomes?” by Douglas Harris and Matthew Larsen. It showed that New Orleans test score growth increased up by more than .2 standard deviations between 2007 and 2010. This was the time, however, when its reformers had even greater freedom in terms of suspending and pushing out students who interfered with their mission to dramatically raise test scores. They also had thousands of additional dollars, per student. Growth then slowed and the next two years’ test gains were almost the same as the two years preceding the hurricane, about .1 standard deviation.


“In other words, nearly a decade of expensive, brass-knuckled reward and punishment produced three years where test score growth was higher than the time when New Orleans was dismissed as a failed school system. NOLA focused completely on raising bubble-in scores, which may or may not indicate that learning increased during that brief window. Harris hopes that better accountability will permanently stop the abuses that proliferated during the time when test score growth increased, but he repeatedly acknowledges doubts that what he sees as effective in New Orleans can be scaled up.

“I hope that Leonhardt will also consider NOLA’s continuing abuses, such as those recently documented by Martha Jewson, and ask whether it will meet the December 2017 deadline for obeying the law.


“The second study cited by Leonhardt claims that charter students in Florida and Chicago did not perform higher in school but had better longterm, out-of-school outcomes. Of course, scholars would have to study hundreds of thousands of students, controlled as best as possible for demographic differences, in order to show that the subsequent increased earnings were a result of charters’ inputs …

Charter High Schools’ Effects on Long-Term Attainment and Earnings (Journal Article)

“Actually, Leonhardt links to study with a sample which includes only 262 low-income students, as well as about 111 special education and about 11 English Language Learners!

“Seriously, this study merely compared students in Florida and Chicago who attended both 8th grade and high school charters in the late 1990s with students who attended 8th grade charters but traditional high schools! The published paper recognized that the small treatment group of 1141 students could be skewed by students not continuing in charters due to discipline problems or family crises. The non-educators who conducted the study ran a series of complex controls that would have been fascinating in a paper on economic theory but that are useless in terms of answering the real world question of whether charters can be more effective in increasing lifetime earnings than traditional neighborhood schools.

“And that leads to the third source which impresses Leonhardt. He cites research by CREDO, but he doesn’t refer to Learning from the Federal Market-based Reforms: Lessons for the Every Student Succeeds Act (ESSA), edited by William Mathis and Tina Trujillo. This anthology acknowledges that CREDO is more professional than “piles of these pseudo-studies/evaluations” by charter advocates, but it still has flaws. Mathis’ and Trujillo’s collection, which Leonhardt doesn’t cite, makes an impressive case that, despite CREDO’s spin, those who hope that charters will close the achievement gap will be disappointed.

Click to access CMO%20FINAL.pdf


“Fourth, Leonhardt links to his editorial in support of Boston’s Match charter schools. His faith in such charters seems to ignore a crucial distinction. Charters may have an “attrition rate” that is no worse than neighborhood schools, but that ignores the “backfill” rate. Charters that don’t fill seats that are emptied are very different than schools that serve everyone who walks through the door, regardless of the time of year.


“More importantly, Leonhardt links to a study which supposedly supports the claim that charters don’t damage neighborhood schools by draining resources and leaving behind greater concentrations of children from generational poverty who have endured multiple traumas. A huge body of journalism and qualitative research, as well as the professional judgments of virtually every teacher who I have ever met, argues for the common sense conclusion that charters have hurt traditional public schools. It would be wrong for anti-charter advocates to ignore the data-driven studies that challenge our conventional wisdom. But isn’t it just as wrong for Big Data researchers to ignore the real world evidence that contradicts their few findings?


“Leonhardt trusts a meta-analysis which concluded that 6 studies showed positive or mixed positive results, with 9 showing neutral results, but with only 1 showing mixed negative and neutral effects. But he doesn’t mention “The Impacts of School Choice Reforms on Student Achievement,” by Gary Miron and Jessica Urschel, which is included in Learning from the Federal Market-based Reforms. In contrast to the studies read by Leonhardt, Miron and Urschel show that 30 charter studies found positive results, with 30 showing comparable negative results, and with 23 showing mixed results.

“I frequently reach out to charter supporters. My friends who say I’m naive for continuing to communicate with the true believers may be right. But, rarely do I find a charter supporter who isn’t disappointed in their outcomes. I doubt that a close reading of the research cited by Leonhardt will find evidence that the flaws in the charter model can be patched up so that they can be scaled up. That is not my big concern, however.

“I hope that Leonhardt and other choice supporters will look anew at the damage done by charters to traditional public schools. Unless they believe that we teachers and our students are suffering from a mass hallucination, its hard to understand how they could use such thin evidence to deny that the additional stress of high stakes testing and increased segregation, both worsened by charters, hasn’t damaged kids, especially hurting our most vulnerable kids.”