Poverty is the single best predictor of poor academic performance, according to a Department of Education analysis of how schools are performing and applying their funding under the weighted student formula.

In a recent briefing to a Board of Education committee, education officials said the latest state assessment showed that 35 percent of students from economically disadvantaged backgrounds were proficient in reading on the 2006 assessment, compared with 57 percent of students who are not poor.

Likewise, in math, 17 percent of economically disadvantaged students were proficient on the 2006 assessment, compared with 34 percent of their more fortunate peers.

"This is not unique to Hawai'i, but it's our problem as well," said Bob Campbell, director of the department's program support and development office.

"We're confirming what people on the Mainland are talking about. The more economically disadvantaged students at a school, the lower the school's reading performance."

State Superintendent Pat Hamamoto told the committee that "poverty is a challenge in teaching" and economically disadvantaged students may not come to class as well prepared or have the same support from home that more affluent students receive.

"Poverty makes a difference and in the DOE we can't solve it alone," said Hamamoto.

New board member Donna Ikeda called on the department to look at addressing some of the effects of poverty by making sure students receive after-school help and other assistance.

Hamamoto said the department has a series of partnerships with entities such as Kamehameha Schools, the state Department of Human Services, the National Guard and other military units as well as nonprofit agencies to provide support for children in high poverty areas.

"They're not as large as we'd like, but there are a lot more than five or six years ago," she said, adding that the department wants to build on the partnerships.

The Legislature also gave the department $800,000 last year to provide direct assistance to middle schools, she said.

Campbell also told the committee that the data are beginning to show there is no correlation between school size and student reading proficiency.

While there's a common belief that students do better in small school environments, the data show that schools with smaller enrollments often face the challenge of being located in high poverty areas such as urban environments, and, as a result, may show lower scores.

"Our small schools have a higher percentage of economically disadvantaged students," said Campbell.

As the number of economically disadvantaged students increases at a school, he also told board members that the number of highly qualified teachers goes down. What that suggests, he said, is looking at incentives to keep teachers where they're most needed.

"This could lead to discussions in that direction," he said.

National studies have shown that schools with a higher percentage of highly qualified teachers — those meeting state certification standards — have students with higher scores.

The ongoing DOE study has also shown there's no uniformity in the workload for staff throughout the public school system, which means there's a wide range in resources schools are offering on a per student basis.

The DOE is attempting to determine the cost of successfully educating one regular education student — and the additional costs to educate students who need extra support.

The hope is to be able to offer more concrete guidelines to future Committees on Weights as its members wrestle with how to divide money among schools according to student need.

Overall, there were 13,253 paid positions in the 2005-06 school year and 13,245 paid positions in the 2006-07 school year, eight fewer positions this year than last.

Reach Beverly Creamer at bcreamer@honoluluadvertiser.com.