Back in April, Andrew Kelly discussed the possible emergence of charter school networks devoted to serving middle class students. To my surprise, however, Kelly didn’t devote much time to addressing the inevitable political objections, e.g., that such networks would represent a pernicious form of “cream-skimming” that would undermine neighborhood schools. The problem with this critique, of course, is that the advent of such charter networks might itself create a situation in which there is more cream to skim, i.e., middle class families will be more likely to remain in cities rather than settle in suburbs as their children start attending school if there are high-quality public educational alternatives to low-performing neighborhood schools. While this does raise legitimate concerns about class segregation, districts could create incentives for high-performing charter networks to serve a more diverse array of students. Moreover, there is a legitimate case to be made — and this is explicit it Kelly’s post — that different charter networks should specialize in serving different student populations. The celebrated KIPP approach has been built around the needs of poor students, many of whom come from disrupted households. Yet students from more stable and affluent home environments might not profit as much from KIPP’s intense emphasis on discipline and attentiveness.
This is a very serious tension. The cliche is that the KIPPs of the world are designed to impart noncognitive skills that middle class children acquire via other channels, and the hope is that students from both backgrounds will be able to come together on a (more) level playing field at some point in the future — in college or in the workplace. Bringing students from such disparate backgrounds together might have some advantages, particularly for the less affluent students: it will tend to reduce the social isolation often associated with poverty and family disruption, etc. Yet it might also undermine teacher effectiveness, as it is difficult to teach across students with extremely varied educational needs and cultural backgrounds. To be sure, we could have class integration at the level of the school rather than the level of the classroom, but this is the kind of tracking that has traditionally served poor students very poorly. Customized learning and blended learning, in which technology and labor are combined to tailor instruction to the needs of individual students, could help address these problems, but my sense is that specialization at the institutional level might have considerable virtues.