Nationwide, a higher percentage of students in established charter schools are judged proficient on the state reading and math examinations than are those in the nearest traditional public school. If a charter school has been operating for more than nine years, 10 percent more students are scoring at or above the proficiency level in both subjects than their public school peers.

This finding comes from a new study by Caroline Hoxby, a member of Hoover Institution's Koret Task Force on K-12 Education and a professor of economics at Harvard University. The study compares the performance of charter school students with students in the nearest traditional public school. Ninety-nine percent of all elementary students in charter schools are included in the study.

For charter schools in operation from five to eight years, Hoxby finds that 5 percent more students reach proficiency in reading than their public school peers. The advantage in math is 4 percent. For charter schools that have been in operation from one to four years, the advantage in reading is 2.5 percent. Overall, 5 percent of charter school students are more likely to be proficient in reading and 3 percent are more likely to be proficient in math on their state's exams than their public school counterparts. (All differences are statistically significant.)

"People involved with charter schools tell you that they improve a lot each year," Hoxby said. "In their first years in operation, they might cope with issues like finding a building. But, after that, they focus on achievement and it shows. Also, low-performing charter schools never attract many students and usually exit quickly. It's the intent of charter school laws: schools that don't work for students should not stay in business."

Hoxby also finds that students in charter schools that receive at least 40 percent of local public school funding do particularly well on the state exams. Additionally, charter schools in areas with a high percentage of poor or Hispanic students appear to provide a special advantage for their students, as compared to the neighboring public school.

"Charter schools tend to arise in areas where students are disadvantaged and families have had little ability to exit underperforming schools," Hoxby said. "Sure enough, charter schools make the most difference with such students."

By adjusting for schools that serve at-risk students, the study focuses on regular charter schools that are expected to meet the same standards as traditional public schools. The study's "matching" method compared charter schools to public schools that are likely to share the same neighborhood, the same economic conditions, and the same population of students and parents. The selection of a neighboring public school ensured that the groups of students being compared were as similar as possible. It is also likely that the public school selected for comparison was the school that most of the charter school students would have attended had there been no charter school.

Caroline Hoxby is a professor of economics at Harvard University, a faculty affiliate in the Program on Education Policy and Governance, the director of the Economics of Education Program at the National Bureau of Economic Research, a member of the Koret Task Force on K-12 Education, and a distinguished visiting fellow of the Hoover Institution. The research is funded by a grant (R29HD35983) from the National Institute for Child Health and Human Development.

A copy of the report is available from the Program on Education Policy and Governance at the following Web site: http://www.ksg.harvard.edu/pepg/