Harvard Study Finds Charter Schools Outperform Public Schools
by Jim Brown
September 17, 2004
(AgapePress) - A new study indicates that charter school students tend to perform better on reading and math tests than their counterparts in traditional public schools in America.In the study, Harvard education economist Professor Caroline Hoxby looked at nearly 100 percent of third, fourth, and fifth grade students in U.S. charter schools. She found that those students did 5 percentiles better in reading and 3 percentiles better in math, as compared with students in the regular public schools the charter students would probably have attended otherwise.
The study flies in the face of a recent report by the American Federation of Teachers that claims charter school students often do worse on student achievement tests than do comparable public school students. Hoxby's study found that, on the contrary, charter school students are more likely to be proficient in reading and mathematics than students of similar racial background in the nearest government school.
The Harvard professor says she is not surprised by these findings since, naturally, no child "is forced to go to a charter school in the United States." She points out that a parent who decides to send his or her child to such an institution, but who feels after a while that the child is not learning or performing as well as he or she should be doing, "will probably not keep the child in a charter school."
Hoxby expects that most mothers and fathers who choose a specific educational alternative for their children are likely to exercise their prerogative again if they do not find their initial choice has served the child's best interests. Noting that educational facilities which are not federally funded might have to excel to survive, she says, "I think that's one of the key things to remember about charter schools -- they don't succeed if they can't keep attracting parents."
The education economist explains the discrepancy between her study and the one by the American Federation of Teachers as the result of a flaw in the AFT's research methods. She says the Federation's study is flawed largely because it failed to make use of a significantly large sample of charter school students.
According to Hoxby, charter schools enroll about one and a half percent of children in the United States. "So if you take three percent of one and a half percent," she says, "you have hardly any students in the sample." She contends that the AFT's previous studies comparing public school and charter school student performance simply looked at too few charter school children to yield accurate results.
"For instance," the Harvard professor says, "in the state of Connecticut, there were only four students in the sample. Or in the State of New York, which is a big state, there were only 32 students in the sample. So basically, they were trying to judge New York's entire charter school policy with about one classroom's worth of students."
Another problem, Hoxby notes, is that the AFT's research did not compare "apples to apples," but instead drew comparisons between children who are demographically dissimilar. For instance, she says the AFT's study compared charter school students in poor urban areas to public school students in affluent suburban districts.
