Charter Schools: Where They Work, Where They Don’t

Supporting charter schools requires tough love. It isn’t enough to create them and let kids attend them. They also need to be run with integrity; their books need to balance; their pupils must be safe; and above all, their academic achievement has to be strong, especially when gauged by student growth.

ednext-may2016-blog-chartersSome of America’s highest-achieving schools are charters, but so are some of its worst. Averaging across all 6,800 of them, some critics declare that their performance is roughly equal to their district counterparts. But such a superficial analysis ignores their variability—the reality that they range from dismal to superb. Let’s look a little more closely.

State variance

A quarter-century in, charter schools are still absent from seven states, and seventeen other jurisdictions have fewer than fifty each. Forty-four states have charter-enabling laws on the books, but these differ so widely that what is possible in one state cannot be accomplished across the border. So it’s no surprise that charter performance differs widely by state. Stanford’s Center for Research on Education Outcomes (CREDO) reported in 2013 that between 2006–07 and 2010–11, charters’ impact on student reading and math achievement varied by as much as 245 learning days per year depending on where the schools were. In Rhode Island, for example, charter pupils gained 108 days in math and eighty-six in reading compared to similar district students. In Nevada, on the other hand, charter goers lost 137 days in math and 108 in reading

School type

Performance also varies widely by school type. “No-excuses” charters, for example, are characterized by high behavioral and academic expectations for pupils, longer school days and years, curricula geared toward college entry, and robust school cultures. They accept “no excuses” for failure, either by children, teachers, or the schools themselves. The best of them—including KIPP, Success Academies, YES Prep, Achievement First, and Uncommon Schools—have done an extraordinary job educating inner-city children, as well as replicating their strategies in networks that can exceed a hundred schools.

KIPP, for instance, has 183 charters serving seventy thousand kids across twenty states and the District of Columbia. Eighty-seven percent of KIPPsters come from low-income families, yet a majority outperform the national average for annual growth across all grades and subjects. Their alumni are 4–5 times more likely than similar peers to graduate from college with bachelor’s degrees. And the Success Academies network in New York City, led by the formidable Eva Moskowitz, now consists of thirty-four schools in four boroughs. When the Empire State adopted the Common Core’s rigorous academic standards in 2014, 64 percent of Success Academy students met them in English language arts—versus 29 percent citywide. Nine in ten were proficient in math, three times the rate across Gotham.

By contrast, “virtual” charters generally yield bleak results. The U.S. contains more than three hundred such schools across twenty-six states, enrolling more than two hundred thousand students. CREDO’s recent analysis shows that, on average, virtual charter pupils achieve 180 fewer days of learning in math and seventy-two fewer days in reading each year than do similar students in district schools. A brand-new study by Civic Enterprise shows that virtual charters also lag far behind in graduation rates.

Community and race

Today’s charters appear to do their best work with disadvantaged youngsters. As University of Michigan economist Susan Dynarkski recently wrote:

In urban areas, where students are overwhelmingly low-achieving, poor, and non-white, charter schools tend to do better than other public schools in improving student achievement. By contrast, outside of urban areas, where students tend to be white and middle class, charters do no better and sometimes do worse than public schools.

A 2015 CREDO study of charter performance in forty-one cities found black and Hispanic students gaining as much as thirty-six and twenty-six days of learning in math and reading, respectively. Asian kids, however, saw no gain. And white students might have been better off in other schools: Those enrolled in charters lost thirty-six days of learning in math and fourteen in reading.

One source of this disparity is school type, discussed above. “No-excuses” schools tend to perform very well, at least when it comes to achievement growth. They also tend to enroll students of color at disproportionately high rates. KIPP’s student body, for instance, is 57 percent black and 39 percent Hispanic. And 93 percent of kids enrolled in Success Academies are children of color.

Meanwhile, more “progressive” white families are likely choosing charters with missions that reflect parents’ dislike of traditional curricula and standardized tests. It may be that their kids are doing OK and benefiting in other ways.

Years in school 

Children’s learning seems to accelerate the longer they stay in charter schools. (Perhaps that’s true of all schools.)  CREDO studies indicate that during a charter attendee’s first year, he or she gains, on average, seven days of learning in math—but loses seven days in reading. In year two, however, charter pupils show positive impacts in both subjects: forty-eight more days in math and forty-three in reading. By year four, the gain is 108 days in math and seventy-two in reading. To get the most from charter schooling, it’s pretty clear that the school has to endure and the student has to stick around!


Oversimplified measures of charter impact communicate none of these nuances, which is why they often lead to the misleading conclusion that charters are no better on average than district-operated schools. Parents might wrongly conclude there’s no reason to move their kids from troubled district schools. And lawmakers may mistakenly decide that supporting charters isn’t worth the political hassle. Yet neither judgment is justified. By no means are weak charters places to celebrate, sustain, or attend. But excellent charters can work wonders—especially for kids who need them most. School leaders, charter advocates, and policy makers just have to learn which models work, why they’re effective, and whom they benefit. We’ll tackle these topics and more in later installments.

– Chester E. Finn, Jr and Brandon Wright

This first appeared on Flypaper. It is the second in a series of essays marking the twenty-fifth anniversary of America’s first charter school law. These commentaries are informed and inspired by the authors’ forthcoming book (co-authored with Bruno V. Manno), Charter Schools at the Crossroads: Predicaments, Paradoxes, Possibilities, to be published in October by Harvard Education Press. They kicked off the series by paying homage to the trailblazers of chartering. Here they address its diverse models and their varying success.

Last Updated


Notify Me When Education Next

Posts a Big Story

Business + Editorial Office

Program on Education Policy and Governance
Harvard Kennedy School
79 JFK Street, Cambridge, MA 02138
Phone (617) 496-5488
Fax (617) 496-4428

For subscription service to the printed journal
Phone (617) 496-5488

Copyright © 2024 President & Fellows of Harvard College