UPDATED
Charter school students are outpacing their peers in regular public school districts in reading and performing at about the same level as traditional public school students in math, according to .
The study, which analyzes charter school performance in 25 states, the District of Columbia, and New York City, found that students attending charter schools gain an additional eight days of learning in reading over the course of a year compared to regular public school students. Charter school students experience about the same amount of learning gains in math as their regular public school peers do.
Both findings indicate an upward trend in performance for charter school students, when compared with the research center鈥檚 , which looked at charter school student performance in 16 states and found that those schools lagged behind their regular public school counterparts in both reading and math.
In the 2009 study, 17 percent of charter schools outperformed their regular public school counterparts in math, about half showed no difference, and more than a third (37 percent) performed worse. In contrast, the new study shows that a quarter of charter schools outperformed regular public school districts in reading, and 29 percent of charters outperformed their local districts in math, while only 19 percent performed significantly worse in reading and 31 percent performed significantly worse in math. That shift represents a significant turnaround for charter school performance compared to the showing of students in regular public schools.
鈥淭here鈥檚 been slow and steady progress in the charter schools, but there still remains more to be done,鈥� said Devora Davis, a research manager at CREDO, in an interview.
The report noted that there are significant numbers of charter schools that are 鈥渆ither substantially worse than the local alternative or are insufficient to give their students the academic preparation they need to continue their education or be successful in the workforce.鈥�
Progress Through Closures?
The study also looked specifically at the 16 states studied in the 2009 report and found that overall, charters in those states experienced academic gains in both reading and math. Researchers attribute those gains to both the closures of low-performing charter schools as well as an overall decline in performance of the regular public schools.
Another difference researchers saw since the 2009 study was that charter schools are educating more disadvantaged students than they were four years ago.
More than half of the charter school population鈥�54 percent鈥攍ive in poverty, a higher percentage than reported in the 2009 study. Charters are also educating more Hispanic students, although they still enroll a lower number of white and Hispanic students than regular public school districts. Charters enroll a higher percentage of African-American students than do regular public school districts.
More must be done to and replicate the high performance of successful charters, the study says.
鈥淲e really want the positive performance in that growing number of charters to be looked at and find out how they can transfer what they do in terms of operating and instructional skills to other charters and public education in general,鈥� said Davis, of CREDO. 鈥淲e really think it鈥檚 time to start and do more of that transfer.鈥�
Andy Smarick, a partner at Bellwether Education Partners, a Washington, D.C.-based nonprofit, said the study was evidence of the charter sector鈥檚 ability to constantly improve, particularly in certain states. Bellwether is a research and consulting organization, whose clients include charters, districts, and states.
鈥淐hartering is a continuous improvement system,鈥� he said. 鈥淚f you do the process right, you鈥檙e going to see this kind of systemic effect.鈥�
Some of the most valuable information from the study comes from the differences shown in charters鈥� performance by state, Smarick added. A common feature of the jurisdictions that fare well in the study鈥攈e cited the District of Columbia as an example鈥攊s that they have strong, independent authorizers, and do not simply 鈥渓eave chartering in the hands of districts.鈥�
Smarick, who has studied urban education, said the results should compel states to eliminate caps on the number of charters.
In the District of Columbia, for instance, where the study found that charters had made significant gains since 2009, 鈥渃harters are approaching 50 percent [of the overall public school population] and getting better results than other schools,鈥� he said. 鈥淚t begs the question 鈥榳hy can鈥檛 we have 75 percent, 90 percent or even 100 percent?鈥� It forces you to ask 鈥榙o we need to keep the district around?鈥欌��
Others were not convinced that the results were a reason for supporters of charter schools to celebrate.
Andy Maul, a fellow at the National Education Policy Center and an assistant professor of research and methodology at the University of Colorado, Boulder, said that overall effect sizes in the study showing differences in academic performance between charters and traditional public schools are 鈥渢rivial in magnitude.鈥� Backers of charters, he said, are 鈥渕aking a lot out of a little.鈥�
The policy center has published numerous reports on charters, some of them critical of either those schools鈥� academic performance, or of the cost of operating high-performing ones.
Davis said charter school authorizers should tie schools鈥� applications to stricter assurances of quality and performance. More research must be done to determine which factors have the greatest impact on charter school performance, she said.
The Stanford researchers drew comparisons between charter school students鈥� and regular public school students鈥� performance through a 鈥渧irtual-control method鈥� in which charter school students were compared to 鈥渧irtual twins鈥� who attend regular public schools where the charter students would otherwise have been enrolled.
Virtual twins are chosen to match charter school students鈥� standardized-test scores, race and ethnicity, special-education considerations, free- or reduced-price lunch participation, English proficiency, grade level, and grade retention in order to provide a comparison between schools with similar characteristics.
鈥業ndistinguishable鈥� Differences?
Maul questioned the methodology of the report, saying that Stanford University researchers could have used more common research techniques, and arguing that the 鈥渧irtual control鈥� model used in the study warranted more scrutiny.
Ultimately, the study will join a 鈥済rowing list of reports that show essentially no overall difference in test scores between students who attend charter and traditional public schools,鈥� Maul told Education Week in an interview.
Similarly, the director of the National Education Policy Center, Kevin Welner, said in an e-mail that the study鈥檚 findings mirror the overall thrust of past research on charters.
鈥淭est-score outcomes for the charter school sector are indistinguishable from the test-score outcomes of the public schools in general,鈥� Welner said.
Policymakers and researchers would be wise to move beyond debates about comparing charters versus regular public schools, and shift to examining 鈥渉ow to improve both of these sectors,鈥� Welner said, 鈥渟tarting with the basic realization that little will be accomplished by merely propping up charters, or by other changes in who runs schools. The most successful charter schools and the most successful non-charter public schools offer greater opportunities, resources and supports for students and teachers.鈥�
Greg Richmond, president of the National Association of Charter School Authorizers, a Chicago-based organization that attempts to improve the quality of charters, said the study鈥檚 state-by-state analysis should allow the public and policymakers to dissect what kinds of policies for governing the sector work best.
鈥淩ather than pretending that the differentiation doesn鈥檛 exist, and proclaiming that there鈥檚 some kind of national truth,鈥� Richmond said, 鈥渨e instead need to say we can now see and learn what鈥檚 happening where [charters are] doing well, and apply those lessons.鈥�
Even in states where charter schools are underperforming, Richmond said he鈥檚 optimistic the study will provide a roadmap for improvement. In some of those states, he said, lawmakers have recently put in place tougher requirements on charters.
Richmond took note that the study found that 8 percent of charters are closing after failing to meet various standards. That鈥檚 a sign that many charters are being held accountable for academic performance, he argued, even if a greater number of ineffective schools need to be shut down.
鈥淐harter schooling is getting better, but we鈥檙e not there yet,鈥� he said.
Sean Meehan and Kevin Connors contributed to this report.