Some charter schools are set up to target specific populations, such as disabled or at-risk students and juvenile delinquents.

The study found the most significant gains in charter school test scores in Florida and Texas. In Texas, charter schools reached math score improvements of 7 percentile points higher than at regular public schools and reading score improvements of 8 percentile points. Florida charter schools achieved one-year math and reading improvements of 6 percentile points each for a student starting at the 50th percentile.

The study listed several reasons why charter schools showed only modest gains over regular public schools: their small size means they usually receive less funding; though they are exempt from many regulations, they still have significant regulatory burdens; and their students are still required to take standardized tests geared to state curricula, removing some freedom to make changes to the curricula.

Another study by the Harvard Civil Rights Project found that 70 percent of Black charter school students attend segregated schools compared with 34 percent of Black public school students. In almost every state studied, the average Black charter school student attended school with a higher percentage of Black students and a lower percentage of White students. Also, there are schools where White charter students are often as isolated as Black charter students. The results are mixed for Latino students (see Black Issues, July 31).