Charter schools are controversial. But are they good for education?

Rigorous research suggests that the answer is yes for an important, underserved group: low-income, nonwhite students in urban areas. These children tend to do better if enrolled in charter schools instead of traditional public schools.

There are exceptions, of course. We can’t predict with certainty that a particular child will do better in a specific charter or traditional public school. Similarly, no doctor can honestly promise a patient she will benefit from a treatment.

Social scientists, like medical researchers, can confirm only whether, on average, a given treatment is beneficial for a given population. Not all charter schools are outstanding: In the suburbs, for example, the evidence is that they do no better than traditional public schools. But they have been shown to improve the education of disadvantaged children at scale, in multiple cities, over many years.

Charter schools are publicly funded but not bound by many of the rules that constrain traditional public schools.