Study: Charters Outperform Public Schools

Published November 16, 2011

Children attending charter elementary schools do better in reading and math on average than those in traditional public schools, according to a University of Washington study of the highest-quality research available. Students at charter middle schools also outperform their traditional counterparts in math.

The study authors, economists Julian Betts and Emily Tang, reviewed 40 studies of charter school achievement and scientifically collated the results.

“If you look at elementary and middle school math results, attending a charter might move a student from the 50th percentile to a 52nd percentile over a year,” Betts said. “That would add up over time. On average, the effects are modest and positive.” 

‘Most Rigorous’ Research
The number of rigorous charter studies has increased in recent years. Betts and Tang considered those randomizing students studied through lotteries and accounting for a student’s history of achievement using value-added comparisons—research considered the “most rigorous” by scientific standards.

The study is an updated version of one Betts and Tang authored in 2008. Betts said the evidence is now stronger that charter school students overall have higher achievement than comparable traditional public-school students.

“It could be that charter schools are getting a little bit better on average over time,” Betts said.

In high schools, little evidence confirms stronger achievement, Betts said, but the results vary by location. 

‘R&D’ Labs of Education
“At our center we’ve had a number of folks working on the collection of what is the right way to judge charter school performance,” said Robin Lake, associate director of the Center on Reinventing Public Education, which commissioned the study. “We convened a consensus panel of the best experts in the country to set standards.”

Lake said the constantly changing charter school research landscape makes regular and updated studies important to determining the schools’ success.

“I think we’re learning a lot from charter schools,” Lake said. “They’re turning out to be what people intended—the research and development arm of public education.”

KIPP Successes Found
Betts and Tang also released results of their study on a national network of open-enrollment, college-preparatory schools called the Knowledge Is Power Program. They determined KIPP schools cause statistically relevant improvement in student achievement: “These impressive effect sizes are enough to move a student initially at the 50th percentile to the 54th and 59th percentiles in a single year.”

Though much of the current research on charter schools has been commissioned in response to controversy, Lake said, the increase in studies is a good thing. The large variation in results between some studies is likely due to the variation in charter school laws and providers from state to state and municipality to municipality.

“Charter schools have to be understood in a local context because state laws vary so much and implementation varies so much,” she said.

‘More Strongly in Favor’ of Charters
Though the 2011 results did not differ much from previous studies, the “overall average effects are more strongly in favor of charter schools than in the earlier review,” the study states.

Many previous studies used “snapshot” approaches that measure data at one point in time, which have been determined ineffective in garnering meaningful results.

“We only found 14 reports that used rigorous methods in 2008,” Betts said. “This time we found 40, so they’re more geographically diverse and cover a greater audience than before.”

Another major difference from the 2008 study was Betts and Tang’s estimate of overall average effects.

“Our new analysis tells us the overall effect and whether it’s statistically meaningful,” Betts said. “We also recorded how much variation we see across studies and if they have a negative impact here and a positive impact there. About three-quarters of the variation across the study is likely to be real variations rather than just lack of precision and uncertainty across estimates.”

Bigger Gap in Urban Districts
The study detected a wider performance gap between charter and traditional public schools in urban areas, with charters much better at boosting student achievement than urban public schools.

“One possibility is that charters operating in urban areas are providing a better education,” Betts said. “A second possibility is traditional schools may be doing a little bit worse in urban areas.”

Discovering which charters are outperforming traditional public schools and why is critical to improving education, Betts said.

“Charter schools were intended to really be hotbeds of experimentation,” Betts said. “They’re free from any state’s education code, which allows them to innovate in terms of curriculum and who they hire as teachers. Part of the promise of the charter school movement is improvement.”

Policy Implications
Local officials should expect “a teething phase” when new charters start, but they shouldn’t be afraid to close down persistently failing charters, Betts said. The relative ease of closing a failing charter school compared to a traditional public school means trouble spots can be more quickly removed from a state’s charter portfolio, she said.

The study recommends charter school “successes should be identified [and] copied in other settings.”

“We’ve had at least three presidents in a row now that look at charter schools as agents of change in the country,” Betts said. “It’s enshrined in our law that charters are an option for improvement.” 

Internet Info:
“The Effect of Charter Schools on Student Achievement,” Julian Betts and Y. Emily Tang, October 2011:


Image by Christopher Michel