Arnold Kling  

Segregation Equilibrium

PRINT
Welcome Guest Blogger Eric Cra... Kiwiana...

From a summary of a study by Donald Haurin and David Brasington:


The study of Ohio school districts showed that an increase of about 20 percentage points in the proficiency test “pass rate” increased house values in a district about 7 percent, even after taking into account other factors that impact house values.

...A student from a privileged background, in a high-income school district, may arrive at school well-prepared and start out scoring well on standardized tests. Years of schooling may not improve that student’s scores. “That school district will look good on average test scores no matter what it does with its students. And its high rating may not be deserved,” Haurin said.

On the other hand, a disadvantaged student in a different school district could end up improving his test scores more than the privileged student, all because he went to a high-quality school. But in the end, if his test scores are not as high as that of the privileged student, the school will not get as much credit, at least in terms of house prices.

“So you can’t look only at proficiency test scores as an outcome and say that is a measure of school quality,” Haurin said. “But that’s what homebuyers in our study did when they were looking for houses.”


Thanks to Nick Schulz for pointing me to Randall Parker's post on the study.

I think that what the researchers are observing is a segregation equilibrium. Affluent parents want their children to go to schools with other affluent parents, regardless of school quality. This may be a status issue. Alternatively, it could be rational. If peer influence is the most important characteristic of a school, and you want your kids to do well in school, maybe it is best to send them to a school where kids generally try to do well in school.

In any case, segregation equilibrium is a challenge for either a voucher system or our current system. If parents will pay extra, in terms of tuition or house prices, to avoid having their children go to school with relatively poor children, then it is hard for poor children to get into schools with achievement-oriented peers.


Comments and Sharing





COMMENTS (7 to date)

Obvious comment: parents aren't interested in the quality of the education of their children, they just want their children to be around similarly advantaged youth so they are safe and have fun at school.

I see this as very similar to how patients choose hospitals. They don't really care so much about actual medical outcomes, they prefer cleanliness and pleasant staff. Why should schooling be any different? I find it a little disheartening in both cases, but ultimately people make their own decisions about what is important. It's too easy to say that people are making "wrong" choices.

Steve Sailer writes:

Right, peer group matters a huge amount in high school and middle school, probably more than the quality of instruction.

Parents are rational to invest lots of their net worths into homes in exclusive public school districts. That's why statewide voucher programs that eliminate the exclusivity of suburban school districts are political suicide, as GOP glamor-boy Bret Schundler found when he ran for governor of New Jersey in 2002 on a voucher platform -- the suburban home-owning moderates that were crucial to his campaign turned heavily against him because his plan would destroy part of their property values.

Vouchers are more likely to be politically feasible when limited to the many urban school districts that already been abandoned by white and Asian home owners.

Steve Sailer writes:

Right, peer group matters a huge amount in high school and middle school, probably more than the quality of instruction.

Parents are rational to invest lots of their net worths into homes in exclusive public school districts. That's why statewide voucher programs that eliminate the exclusivity of suburban school districts are political suicide, as GOP glamor-boy Bret Schundler found when he ran for governor of New Jersey in 2002 on a voucher platform -- the suburban home-owning moderates that were crucial to his campaign turned heavily against him because his plan would destroy part of their property values.

Vouchers are more likely to be politically feasible when limited to the many urban school districts that already been abandoned by white and Asian home owners.

Mark Seecof writes:

I think, in fact, that parents want to send their children to schools in which all of the other kids are bright and well behaved. To this end they look at the income of the other kids' parents as a proxy for the qualities desired in the kids. I don't think most parents would object to bright, well-behaved poor kids in any school, should some happen to show up.

So it would seem that the obvious way to to overcome objections to public-school choice is simply to sort pupils by intelligence, diligence, and good behaviour. "Tracking" within a school is a partial implementation, distinct schools ("Metro Latin Academy" vs. "Southside Elementary") would be a full implementation. However, this approach always fails, for two distinct reasons:

(1) Parents of less bright or less-well-behaved children will not accept the sorting, no matter how just. Parents of dimwits and troublemakers hope that their own kids will do better if surrounded by good examples--or at least, have fewer opportunities to "hang around with slackers" and get into trouble.

More affluent parents work (very effectively) to prevent school-choice programs so they can continue to buy places in better schools for their children.

(2) Sorting (tracking) will reveal ethnic disparities in academic performance. No matter how real these are, no matter how objectively they are measured, any program which reveals them will be denounced as racist. People will sue the school authorities, alleging racial bigotry.

School officials work to prevent school-choice programs for many reasons, but one key reason is so they won't have to make decisions which would be denounced as racist. Schools in communities sorted by income serve pupils pretty-well pre-sorted by race, so staff never make academic decisions which incidentally reveal ethnic disparities.

Together these problems will sink any attempt to provide serious school choice in our current society.

Half Sigma writes:

I'm in agreement with Steve Sailer when he says "peer group matters a huge amount in high school and middle school, probably more than the quality of instruction."

If you send your kid to a school where all the other kids are diligently preparing to attend Ivy League colleges, then your kid will pick up those same attitudes.

If your kid attends a school where the other kids care more about partying and don't value education, then that's the behavior your kid will model themselves after.

You see, parents really DO have a huge amount of influence on their kids, because they get to choose who their kids peers are by moving to the right school district!

dearieme writes:

The vocabulary in which all this is dicussed is very odd. How come going to a high quality school doesn't mean that you are "privileged"?

(-_-) writes:

to Nathan: You are comparing patients to parents. The difference is one is directly effected by the pleasent staff, and the other is directly concerned with only the outcome.

Comments for this entry have been closed
Return to top