Campus & Community

Charter schools get high grades

7 min read

Charter students more proficient than traditional peers

For many parents, educators, and policy-makers in the United States, charter schools – innovative public schools that are free from much bureaucratic

Economics Professor Caroline Hoxby has released a comprehensive study that shows that students at charter schools perform better on standardized tests than their peers at comparable traditional public schools. (Staff photo Stephanie Mitchell/Harvard News Office)

oversight but must “compete” for students in order to retain their charters – have held out enormous promise as a public alternative to failing traditional schools. So when the American Federation of Teachers (AFT), the nation’s second-largest teachers’ union, published a study in August 2004 that found students at charter schools performing worse than their peers at traditional public schools, more than a few hopes were dashed.

But to Harvard Professor of Economics Caroline Hoxby, something in the much-publicized study seemed amiss. With her economist’s logic, she wondered why parents would keep their children in charter schools, which are a breeze to walk away from, if their kids were performing poorly?

Hoxby quickly gathered data, and just a month after the AFT study grabbed headlines, hers – “A Straightforward Comparison of Charter Schools and Regular Public Schools in the United States” – is making news with findings showing that, on average, students in charter schools are 5 percent more likely to be proficient in reading and 3 percent more likely to be proficient in math than students at the closest public schools with similar racial composition. In states where charter schools are well established, the advantage is even more pronounced – up to 35 percent greater proficiency among charter school students in the District of Columbia, for instance.

Hoxby maintains her large sample size makes her findings robust and, ultimately, more credible. She used standardized test results from 99 percent of the nation’s charter schools, while the AFT study that prompted hers surveyed

See Professor Hoxby’s study

similar results from only 3 percent of charter schools. “You can’t learn about charter schools by studying 3 percent of kids in charter schools, because charter schools students are only one and a half percent of American students,” says Hoxby. In Connecticut, for example, that translates into just four students. “You have to have more students in a study before it can deliver serious results,” she says.

Comparing apples to apples

For all their bureaucratic freedom, charter schools are required to administer the same standardized tests as traditional public schools. Hoxby accessed state examination scores for fourth-graders in charter schools. In what she calls a very important effort to compare apples to apples, she looked at the same test scores from the school the charter students’ would most likely otherwise attend: either the nearest regular public school or the nearest traditional school with a similar racial composition.

“It would not really be helpful for me to take a kid who’s going to an inner-city Boston charter school and say, ‘What if he went instead to the Weston public schools?’” Hoxby says, citing one of Massachusetts’ best-performing and wealthiest suburban school districts to point out a flaw in any attempt to compare charter schools, which are very often located either in inner cities or rural areas, with American public schools in general.

From a policy perspective, Hoxby says that the even stronger proficiency boost in states with bigger, more long-standing charter school movements – Arizona, California, Colorado, and the District of Columbia among them – is noteworthy. “I don’t think we’re going to learn a lot by looking at states with only six charter schools that started last year,” she says, noting that in their first year or two, charter schools can be “oddball” places, operating out of makeshift facilities and populated by students whose parents are either very experimental or desperate to improve their child’s failing performance.

Charter schools in existence for five years or so have a track record (if not, in theory, they would have failed). These schools, says Hoxby, can tell us much more about the potential of charter schools to improve students’ performance.

Perform or go out of business

Hoxby chalks up charter schools’ success to the old-fashioned free market competition that governs them. “On average, they’re doing better because if they’re not well managed, they go out of business,” she says. Yet the ways in which they’re better at serving students are far from bean counting.

While viewing parents as customers who are free to “shop” at the traditional school down the street may seem harsh, Hoxby says it has forced charter schools to set up a different – and positive – model of parent interaction. “Parents can walk away at any time, and I think this changes the whole nature of the relationship between parents and schools,” says Hoxby. “A lot of parents who send their kids to charter schools have never dealt with a school that cared about their opinions before. It’s really wonderful how engaged they get when they see that the school needs them. It’s inspiring.”

And because they’re under constant scrutiny, successful charter schools tend to use proven systems of curriculum and pedagogy. “If you’re a charter school and your students aren’t learning to read, you’re probably not going to stay in business,” she says. “Parents have reasonably good judgment about their kids. … If you send your child to a charter school and you feel that things aren’t going that well, you take your kids out of the charter school and go back to the regular public school.”

Research, not politics

Hoxby deftly counters common attempts to undermine charter schools’ achievement superiority – for instance, rather than “cream-skimming” the best kids out of public schools, charter schools actually tend to do the opposite, attracting the most troubled kids who are floundering in regular schools. But despite her statistically sunny view of charter schools, they’re generating increasingly polarized and politicized emotions.

Although parents who use them and those who run charter schools are bipartisan and range from the most liberal to extremely conservative, it’s easy to see how the movement has become a political one. The Bush administration’s No Child Left Behind act touts charter schools (although Hoxby argues that they are a minor piece of that legislation in practice), while the AFT, together with the National Education Association, which is the nation’s largest teacher union, comprise the single largest contributor to the Democratic Party.

“I don’t think charter schools should be a polarizing or political issue if you look at them through the lens of kids,” she says. “Charter schools just don’t have anything to do with politics.”

She hopes that policy-makers and the American public will leave the politics out of charter schools and let researchers like herself continue to study them in a rigorous, scientific way. Even more powerful than the research methods that produced this report, she says, are randomized studies – true apples-to-apples comparisons, like one she and a colleague did with Chicago schools – that examine charter school students and those who would have liked to attend a charter school but are in a traditional school.

“Let’s leave charter schools alone until we’ve had enough time to evaluate them scientifically, with the randomized studies that are increasingly available,” she says. “I think we’ll end up with better education in the United States if we treat it seriously.”