A national study suggests one way school districts can improve graduation rates for high school students: get geographically smaller.
The study, released this month by the Manhattan Institute, a think tank based in New York City, draws on federal data to analyze changes in districts鈥� geographic size, high school enrollments, and graduation rates from the early 1990s to 2001. Researchers found that graduation rates increased, for the most part, when districts downsized and decreased when they grew in size. The study was based on federal data for school districts nationwide.
a report analyzing changes in districts鈥� geographic size, high school enrollments, and graduation rates, is available from the .
By the authors鈥� calculations, for example, a district that shrank by 200 square miles might expect to see a 1.7 percent improvement in its graduation rate over a period of several years. For instance, if Florida, a state with unusually large school districts, many bigger than 800 square miles, was somehow able to decrease its districts to about a third of their current geographic size, the change could theoretically increase the state鈥檚 average graduation rate from 59 percent to 64 percent.
Jay P. Greene, the study鈥檚 primary author, said his findings run counter to the conventional wisdom on cost-effectiveness that has prompted many districts to consolidate.
鈥淎s it turns out, there are things about schools that are different from business,鈥� said Mr. Greene, a senior fellow at the institute who is based in Davie, Fla., 鈥渁nd one of the differences is that you don鈥檛 get economies of scale in education.鈥�
He said one reason that downsizing might lead to greater effectiveness is that it can give parents more choice of districts for sending their children to school.
鈥淭he larger a district is, the harder it is for families to leave it,鈥� said Mr. Green, who has done studies linking school choice to improved academic outcomes. 鈥淲hen you have a lot of smaller districts, they鈥檙e competing for students and tax revenue.鈥�
Bit of a Stretch?
But some experts said Mr. Green鈥檚 suggested link between school district size and greater choice may be a bit of a stretch.
鈥淚t鈥檚 really not choice鈥攖hey鈥檙e just talking about geographic size,鈥� said Clive R. Belfield, an associate director of the National Center on Privatization in Education, based at Teachers College, Columbia University, in New York City. He also faulted the study for failing to account for the reasons why districts choose to increase or decrease in geographic size.
Though the median size of districts nationwide is 260 square miles, the Manhattan Institute study shows that states vary widely in that respect. Hawaii, for instance, has just one large state-operated district of 6,460 square miles, while the average Massachusetts district measures just 22 square miles. In recent years, Hawaii has considered breaking its one district into smaller units. Other states, such as Arizona, Illinois, and Arkansas, however, have done, or considered doing, the opposite.
Mr. Greene said his findings suggest the latter option may be unwise.
The Manhattan Institute paper is likely to draw criticism for the manner in which it was released. Like other papers the think tank produces, the district-size study was not widely reviewed by researchers, and it had not been published in a scholarly journal. And some education researchers were quick to note last week that Mr. Greene had signed a full-page advertisement in The New York Times last August that criticized the American Federation of Teachers for doing much the same thing with a study that criticized charter schools.
