« Commendable Second, But Unpersuasive Third, Thoughts On Affirmative Action | Main | Update / Correction »

St. Pete Parents Unconcerned With Schools’ Race Mix

A reader sent a link to a fascinating article in yesterday’s St. Petersburg Times reporting the results of a large survey of parents regarding what was important to them in choosing schools for their kids.

St. Petersburg euphemistically refers to its school assignment policy as a “choice” plan. But note:

In the four years since the choice plan began as a way to voluntarily integrate schools, Pinellas’ busing costs have surged from $26.6-million to $47-million, an increase of 77 percent.

A citizen task force will use the survey as it prepares a final report to the School Board later this month on the future of the choice plan. The plan’s strict race ratios, which cap the number of black students in any school at 42 percent, will expire in May as part of a federal court settlement that brought a gradual end to 30 years of busing. [Emphasis added]

Some choice.

The results of the survey will not surprise many people, except school officials and liberals, who have followed the debates over school assignment. “Even among most black families,” the SPT reports, “race is a distant concern in school selection.”

Far more important to black families, the survey indicates, are factors such as a school’s academic reputation, its proximity to home and the presence of special programs in math, science, art, music and other areas.

Pinellas black families were about five times more likely than whites, and twice as likely as other minorities, to consider the racial makeup of a school’s staff and student body. But when asked to list the top three reasons for choosing a school, those factors proved to be secondary at best.

Only 16 percent of black families put the racial makeup of schools as a top-three concern. Even smaller numbers said that having more black teachers and administrators was that important.

Taken together without racial breakdowns, 68 percent of Pinellas families said the racial makeup of a student body does not impact their choice of schools. More than 75 percent said the racial makeup of a school staff does not factor into their decision

Based on the survey results as well as written comments on many survey forms,
The USF [University of South Florida] researchers who analyzed the survey detected a sense that many Pinellas families had grown weary of categorizing students by race.
I haven’t conducted any surveys, but, surprise: I’ve detected that same sense. And good sense it is, too.

TrackBack

TrackBack URL for this entry:
http://www.discriminations.us/sa/mt-tb.cgi/5994

Say What?

You've got to love the "Even among most black families" line, as if blacks should somehow be less concerned about academics and more concerned about race than whites.

There's so much social desirability bias involved here that it's hard to put much credence in a survey. A better technique would be to look at expressed preference (i.e., all other things equal what racial mix do parents implicitly target when sending their children to school).

But doesn't this assume parents do in fact target racial mix to one degree or another? Are there no parents who care only about other matters (quality of academic program, availability of sports/band etc., distance from home) and don't give a hoot about racial mix? Wouldn't such a survey implicitly exclude such parents, or force them to come up with something that they wouldn't have thought of on their own?

John,

The analysis I was proposing wouldn't stack the deck (and in fact it wouldn't be a survey).
The necessary dataset would be data on which students requested assignment to which schools. At the student level you'd need to know race and home address. At the school level you'd need to know address, racial mix, test scores, athletics, AP classes, and whatever else might make a school desirable.
You'd then use logit or probit regression analysis to see which students requested which schools. Based on this you could see what actually does matter to parents rather than what they'll admit to telling you. For instance, by what they tell you it's probably academic performance but by what they do it's probably geographic proximity. Whether any particular variable (including race) actually matters is an empirical question that is not assumed by the analysis.

Anon - I'm persuaded. Has any analysis like this been done anywhere?

Post a comment