School redistricting never easy, but it shouldn't be this hard

I've always said school redistricting was a thankless job (you can look it up). It has to be done periodically because populations shift inevitably. But no matter how you do it, some will be convinced that officials of the school system don't give a hoot about their kids and are bent upon tearing apart neighborhoods for the sake of their own convenience or worse.

So I can appreciate why Joel Gallihue, the county's manager of school planning, took exception to an editorial we ran recently bemoaning the county's ability to accurately predict the shifts in school populations and urging the county executive to appoint a panel to figure out ways to improve the methodology. Our editorial board had made the same suggestion eight years ago.

"The editorial from June 19, 2003, would have been referring to the accuracy of a projection made in February of 2002 for the school year beginning August 2002. When this projection is compared to actual enrollments, 74 percent of the 66 school projections had error rates at or below 5 percent," Gallihue said in his email to me. "Eight schools had error rates above 10 percent. The system level error across all schools was 4.1 percent (mean absolute percentage error, or MAPE).

"The accuracy for the projection made for last school year was evaluated by the Board of Education on February 10, 2011. Staff reported that for the 71 schools for which we provide annual projections, 82 percent had error rates at or below 5 percent and none exceeded 10 percent. The system level error across all schools was 2.9 percent (MAPE). These data indicate improvement and do not support your position that projections are no more reliable today than they were in 2003."

So these predictions are getting better after all. Can't argue with that. The question then becomes whether they're good enough.

When we spoke on the phone, I told Gallihue that, in retrospect, we were probably tougher on the school system than we meant to be in that editorial. It's clear school officials have their hands full trying to cope with residential growth that, even amid a prolonged and deep economic slump, continues at a pace many U.S. counties would kill for these days.

A big reason for that, naturally, is the reputation Howard County enjoys for having high-quality public schools. In short, the school system is a victim of its own success.

But it's also been a victim of an adequate-public-facilities ordinance that has not been adequate to the task of sufficiently controlling residential development to prevent school overcrowding for any appreciable length of time, even though our little county has opened 29 new schools since 1990.

County school officials say that in order to optimize resources across the system, they try to keep each school's enrollment between 90 and 110 percent of capacity. Currently, 18 county schools have enrollments that exceed 110 percent.

"Once you get past 110 percent, life becomes a little more difficult," Gallihue says. "Once you get to 115 percent, it becomes a lot more difficult."

The 19-year-old county ordinance ensuring that schools aren't overwhelmed doesn't curtail development until an affected school reaches 115 percent. Even then, the school system has just four years to un-crowd the school, either by redistricting or by opening a new school, or the development can go ahead anyway.

And high schools don't factor into the APFO equation. Howard High is one of the county's over-110 schools.

Gallihue notes that enrollment projections got turned on their heads by the economic swoon of the past few years. The boom in the west county fizzled as young families became less ambitious in their home purchases. "It's difficult to get a mortgage these days, let alone one for a giant house in western Howard County," he observes.

So long, mini-mansion; hello, condo.

The turn in the nation's fortunes has been extraordinary, of course, but there's nothing unusual about fights over school redistricting around here. We will never eliminate them, but they shouldn't be as frequent and widespread as they are now. It's time for an APFO overhaul.