I take it that "racially isolated" is a euphemism for "black." In any event, mightn't it be easier to raise salaries in urban school districts (which at some point would surely attract a better teaching force) than to pine away after fully integrated school bodies (which will likely never happen within our lifetimes)?

Debbie and Diane? They seem to avoid getting dragged into the mud quite so easily. Over at Flypaper, Milli Petrilli and Liam Julian headline their postings not once, but twice, “Does Eduwonkette think teachers are racist?” Gee, guys. Does she actually say anything like that? Or, maybe, is she simply summarizing the work of respected economists who have studied teacher turnover in Texas, North Carolina, Georgia and New York, all of whom find that teachers are more likely to leave schools with higher concentrations of minority students? Is Eric Hanushek calling teachers racists? Is Steve Rivkin?

What they, and others, are documenting is that there frequently is an interaction between teacher race and the racial composition of the school, and the exits of white teachers are more responsive to the school’s racial composition than are the transitions of black and Hispanic teachers. And what they say is, “… if schools with high minority concentrations have more disciplinary problems, rigid bureaucracies, poor leadership, high student turnover, and general safety concerns, improvement in such directions may reduce teacher turnover. (And, improvement in these dimensions may simultaneously have a direct benefit for student performance). In addition, improvements in academic preparation, such as through better preschools or child care services, may well have the indirect benefit of making schools more appealing to prospective teachers. Learning more about the precise sources of the relationship between teacher labor supply and the specific student characteristics would provide important, policy relevant information.”

Mike Petrilli’s retort is that schools such as KIPP and Amistad Academy are counterexamples to the claim that good teachers shun racially isolated schools: “Don’t they show that great teachers will teach at racially isolated schools that are well-run?” The jury’s still out on whether these schools are academic wonders populated by great teachers, although that’s certainly a possibility. But even if that were true, these schools represent a tiny fraction of the racially isolated schools in the U.S. Just a few days ago, eduwonkette posted data from the new edition of the Condition of Education. She’s as shocked as I am at the figures. In central cities, 66% of Black students, and 70% of Hispanic students, attend schools that are at least 75% minority students. In contrast, 9% of white students residing in central cities attend schools with at least 75% minority students. Conversely, 69% of white students in central cities attend schools that are at least 50% white, whereas only 14% of Black students and 12% of Hispanic students do. Houston – and L.A., Detroit, Chicago, New York, Boston, Philadelphia and Seattle – we have a problem.

Very well said, skoolboy - although one point you failed to include (a point that Petrilli and Company fail to see time and time again) is that student populations differ in important ways even among racially isolated schools.

Kids in racially isolated KIPP schools, for example, are not otherwise identical to kids in racially isolated schools in the general population, as much as they might appear so from the comfortable boardroom of the Fordham Foundation. By design, they and their parents have chosen to be there.

If teachers are less likely to leave racially isolated choice schools, it is almost certainly because these schools and their students bring something to their job that other schools and their students do not, such as high levels of parental time and commitment, better attendance rates, safety, or student behavior.

So what is your proposed solution?

To have government discriminate on the basis of race to achieve a racial balance you think is more desirable?

Absent past intentional racial discrimination, that seems to have some constitutional infirmities after last year's SC decisions. It also seems to be a policy with an horrendous track record, cf. bussing. Not to mention the fact that there is scant evidence that placing these kids in more racially balanced schools is going to increase their academic achievement. (And by evidence I mean something that passes the laugh test for educational research.)