Matthew DiCarlo at the Shanker Blog is a careful social scientist who does thoughtful analysis of education issues.

His blog today reviewed the GAO report on special education enrollments in charter schools.

The report got lots of attention for finding that about 11% of students in the nation are special education, but charters enroll only 8%.

James Shelton at the U.S. Department of Education was unimpressed by the disparity and basically brushed it off as “a fine point”  and “a relatively small difference” that he would look into one of these days. Of course, given Shelton’s role at the Gates Foundation and Race to the Top, one would not expect him to be exercised about issues in the charter sector. His job is apparently to expand them, not to hold them accountable.

Matthew DiCarlo took a closer look and pointed out that it was misleading to compare national data on special education enrollments to charter data, because so many charters are located in urban districts, where the proportion of special education students is higher than it is nationally.

He takes the example of Ohio, which seems to have the same proportion of special education students in charter schools and in public schools, but this is misleading. When DiCarlo looked at the charter enrollments in the state’s cities, he found wide disparities.

I noticed the same phenomena yesterday when I read about the report and its findings. Last fall, I spoke at an education summit in the Bronx, which is the poorest borough in the city of New York. Most of the borough’s 35 charter schools are clustered in the South Bronx, which is the poorest neighborhood of the poorest borough. The enrollment of special education students in the South Bronx public schools is 19%; in the charter schools in the same area, it is 11%. That is not a small or insignificant difference. Similarly, when I looked at the enrollment of English-language learners, there were nearly twice as many ELLs in the publics as in the charters in an area with many Hispanic families.

There is yet another issue that should be considered when comparing the presence of students with special needs in different sectors, and that is the degree of the students’  disability. Some charter schools take children with mild learning disabilities, but not the children who have severe disabilities. The students who are autistic or need feeding tubes or have other high needs are over-represented in the public schools and under-represented in the charters. That helps the charters show higher test scores and greater gains.

The disparity underlies an important point: Comparing the test scores of public schools and charter schools is an apples and oranges comparison. If they don’t enroll the same children, and they have different challenges, then the comparison is inherently unfair. But there is a larger issue that we should think about. Long-term, what will it mean to education in big cities when the children with the greatest needs are concentrated in the public schools, and those who are most motivated, are the least costly to educate, and have the lowest level of need are clustered in charter schools? Will it mean dual systems, both publicly funded but serving different populations?