Charter Schools Lag, Study Finds

By Jay Mathews
Washington Post Staff Writer
Wednesday, August 23, 2006

Fourth-graders in traditional public schools nationwide did somewhat better on average than those in charter schools in reading and mathematics in 2003, a long-awaited federal report said yesterday.

Earlier versions of the data have been used as weapons in a lively political and academic war between charter school advocates and opponents, but the new National Center for Education Statistics study appeared to provide little new ammunition for either side and little guidance for people trying to judge their schools.

"What does the report say to a parent? Not much, frankly," said Mark Schneider, commissioner of the center, in a telephone news conference yesterday.

The center looked at 6,764 traditional public schools and 150 charter schools, which are public schools that operate independently. It said traditional schools scored 4.2 points higher in reading and 4.7 points higher in math on the 500-point National Assessment of Educational Progress test for fourth-graders, after adjusting for such student characteristics as family income. This was the first time such adjustments had been reported for the 2003 data.

The study emphasized that the results could have been distorted by several factors it could not adjust for, such as the lack of a random sample, different levels of parental support and different levels of learning before the students reached fourth grade.

The Washington Post reported yesterday that the District has 23 percent of its public school students in charter schools, a higher percentage than any other school district in the country. D.C. School Superintendent Clifford B. Janey has called for a moratorium on new charter schools but has received little support from elected officials, who note that voters have very little confidence in the traditional public schools.

Two recent studies show D.C. charters outperforming traditional schools, but they are subject to the same problems of inadequate data and difficult interpretation that the center's report acknowledged in its national study.

"These studies are just snapshots," said Tom Loveless, a senior fellow at the Brookings Institution. "They tell us how students in two kinds of schools are performing at one point in time. Unfortunately, they don't answer the question of which kind of school -- charter or traditional -- is better."

Schneider rejected the suggestion by some reporters that he might be playing down the significance of his report because of the political battle. He said national data could not help a parent trying to compare schools in a local neighborhood because individual schools vary widely in quality. "My advice to parents is shop around carefully," he said.

The report, available at , said that when researchers looked only at schools in cities with high minority populations, the difference in reading scores between the average traditional school and average charter school disappeared.

The study also tried to compare just charter schools associated with a local school system, such as the D.C. charter schools, with traditional public schools. In both reading and math for fourth-graders, there was no significant difference. Charter schools that have no tie to a system, and often draw students from several systems, scored lower than traditional schools.

Several experts said the study had either fortified or failed to shake their view of charters. American Federation of Teachers President Edward J. McElroy said the report "provides further evidence against unchecked expansion of the charter school experiment."

Jeanne Allen, president of the Center for Education Reform, criticized the methodology and said "charter schools are succeeding and do better in most cases."

Chester E. Finn Jr., president of the Thomas B. Fordham Foundation, which supports education research, said the study added little useful to the debate. "Bottom line: big yawn," he said.

© 2006 The Washington Post Company