Students in New York City charter schools are, on average, posting higher gains in reading and mathematics than they would have had they attended the city鈥檚 regular public schools, a concludes.
Issued this week, the report comes amid a continuing national debate on how charters stack up academically. It uses what the authors call the 鈥済old standard鈥� in research: a randomized trial of students who entered lotteries to attend charter schools compared with students who applied but did not win slots. The study finds the strongest charter gains in math.
To help explain the gains, the study couched the growth for charter students in terms of New York鈥檚 system for classifying students鈥� performance levels. The state ranks students in four categories: not meeting learning standards, partially meeting them, meeting the standards, and meeting them with distinction.
A charter student in grades 3-8 is gaining about an extra 12 percent of a performance level in math each year over the comparison group, the study says. In reading, the growth is approximately an extra 3.5 percent each year.
鈥淭his means that a charter school student whom we would have expected to be failing if he had stayed in the traditional public schools would be, at the end of 13 years of charter school education (K-12), above proficient in math,鈥� Caroline M. Hoxby, an economics professor at Harvard University and a co-author of the study, wrote in an e-mail.
More than 90 percent of the charter applicants in the study came from low-income families, far more than in the 1.1 million-student school system as a whole, the authors say. Sixty-four percent were black, compared with 32 percent of students citywide.
The findings were statistically significant in both subjects.
鈥楽lippage Points鈥�
The report, the first out of a multiyear study paid for by the U.S. Department of Education, was based on several years of data.
The authors said the results should be viewed with caution, though, as the data set is limited but will expand quickly. Most students had been in charters, which are public but largely independent schools, for one to three years, Ms. Hoxby said.
Researchers lacked enough data to draw firm conclusions about the factors leading to the higher achievement. But one promising trait, the authors say, is a longer school year.
The report looks at various kinds of data from 42 of New York City鈥檚 47 charter schools, with test results for 35. The study did not have enough data on high school students, but the authors expect it to in future years.
鈥淸This] is one of a small but growing number of studies that use the gold standard of lotteries to evaluate the effect of charters on achievement,鈥� Julian R. Betts, a professor at the University of California, San Diego, said via e-mail.
Mr. Betts, an expert on school choice, called the findings encouraging for charters. 鈥淚t is going to take some time, though, before we really understand whether these convincing gains transfer to the high school level,鈥� he said.
鈥淚t鈥檚 an important study鈥攖here鈥檚 no getting around that,鈥� said Jeffrey R. Henig, a professor of political science and education at Teachers College, Columbia University, who also studies school choice. But he said a lottery-based analysis comes with important caveats, because 鈥渢here are all kinds of slippage points where researchers have to make certain calculations to better approximate what you would have gotten with a fully controlled experiment.鈥�
He added: 鈥淚f these positive results hold up, you鈥檇 have to be very wary about generalizing them to other places where the funding is less generous, the district is less hospitable, and the chartering bodies are different.鈥�
