"This study compares the reading and mathematics proficiency of charter school students in the United States to that of their fellow students in neighboring public schools. Because charter schools are public entities, their students take state exams. Thus, this study is based on schools that enroll approximately 99 percent of elementary students who attend charter schools. The charter schools are compared to the schools that their students would most likely otherwise attend: the nearest regular public school with a similar racial composition (the ‘matched’ school).

"Compared to students in the matched public school, charter students are 5.2 percent more likely to be proficient in reading and 3.2 percent more likely to be proficient in math on their state's exams. Charter schools that have been in operation longer have a greater proficiency advantage over the matched public schools. For example, in reading, the advantage is 2.5 percent for a charter school that has been operating 1 to 4 years, 5.2 percent for a school operating 5 to 8 years, and 10.1 percent for a school operating 9 to 11 years.

"The results show that charter schools are especially likely to raise the achievement of students who are poor or Hispanic. This is a useful finding because charter schools serve students who are disproportionately likely to be minorities or poor."

"In states where charter schools are well-established, charter school students' advantage in proficiency tends to be greater. For instance, in Arizona, fourth grade charter students are about 10 percent more likely to be proficient in reading and math than students in the matched regular public schools. In California, the corresponding proficiency advantages are 9 percent in reading and 5 percent in math. In Colorado, the corresponding proficiency advantages are 12 percent in reading and 14 percent in math."

"Because charter schools enroll only 1.5 percent of students, it is important to include nearly all of them in a study. Results based on only a small sample of charter school students (for instance, studies that rely on the 3 percent sample of the National Assessment of Educational Progress) cannot be used to draw conclusions about states' charter school policies. A study that relies on a 3 percent sample of 1.5 percent of American students is a study based on only 0.045 percent of students. In contrast, this study uses data that are sufficient for detailed investigations of charter school students' proficiency, nationwide."